CN118336829A - Power control method, system and equipment based on source network charge storage - Google Patents
Power control method, system and equipment based on source network charge storage Download PDFInfo
- Publication number
- CN118336829A CN118336829A CN202410311320.2A CN202410311320A CN118336829A CN 118336829 A CN118336829 A CN 118336829A CN 202410311320 A CN202410311320 A CN 202410311320A CN 118336829 A CN118336829 A CN 118336829A
- Authority
- CN
- China
- Prior art keywords
- meteorological
- weather
- power grid
- feature
- sequence
- 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 40
- 238000004458 analytical method Methods 0.000 claims abstract description 143
- 238000013507 mapping Methods 0.000 claims abstract description 43
- 238000012549 training Methods 0.000 claims abstract description 42
- 230000004927 fusion Effects 0.000 claims description 26
- 238000000605 extraction Methods 0.000 claims description 24
- 230000015654 memory Effects 0.000 claims description 19
- 230000006870 function Effects 0.000 claims description 12
- 238000002864 sequence alignment Methods 0.000 claims description 10
- 238000012163 sequencing technique Methods 0.000 claims description 10
- 238000012300 Sequence Analysis Methods 0.000 claims description 6
- 238000001514 detection method Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 125000004122 cyclic group Chemical group 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000004891 communication Methods 0.000 description 10
- 238000013528 artificial neural network Methods 0.000 description 7
- 238000007405 data analysis Methods 0.000 description 6
- 238000007726 management method Methods 0.000 description 5
- 238000010248 power generation Methods 0.000 description 5
- 238000013473 artificial intelligence Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000004146 energy storage Methods 0.000 description 3
- 238000003062 neural network model Methods 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 230000036962 time dependent Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 230000009193 crawling Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007500 overflow downdraw method Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000010937 topological data analysis Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention relates to the technical field of power control, and discloses a power control method, a system and equipment based on source network charge storage, wherein the method comprises the following steps: carrying out power grid structure analysis and positioning modeling on the historical power grid data set to obtain a power grid structure model; performing geographic grid division and meteorological feature clustering mapping operation on the historical meteorological data set to obtain a meteorological grid model; performing time sequence weather training on the weather grid model by using the historical weather data set to obtain a weather analysis model; performing joint training on the power grid structure model by using the historical meteorological data set and the historical power grid data set to obtain a power grid analysis model; and calculating and analyzing the power grid data by using the weather analysis model, the power grid analysis model, the real-time power grid data and the real-time weather data, and controlling the power of the target area according to the analyzed power grid data. The invention can improve the efficiency of power control.
Description
Technical Field
The present invention relates to the field of power control technologies, and in particular, to a power control method, system, and device based on source network load storage.
Background
The energy source network charge storage is a concept integrating energy, a power grid, load and energy storage, and aims to realize efficient utilization of the energy and sustainable development of a system.
The existing power control method of the power grid is mainly based on a power control method of a power grid mathematical model, and mainly comprises the steps of carrying out power modeling on each part of a source network load storage so as to realize power control.
Disclosure of Invention
The invention provides a power control method, a system and equipment based on source network charge storage, and mainly aims to solve the problem of low efficiency in power control.
In order to achieve the above object, the present invention provides a power control method based on source network load storage, including:
Acquiring a historical power grid data set and a historical meteorological data set of a target area, and carrying out power grid structure analysis and positioning modeling on the historical power grid data set to obtain a power grid structure model;
performing geographic grid division and meteorological feature clustering mapping operation on the historical meteorological data set to obtain a meteorological grid model;
Performing time sequence weather training on the weather grid model by using the historical weather data set to obtain a weather analysis model, wherein the performing time sequence weather training on the weather grid model by using the historical weather data set to obtain the weather analysis model comprises the following steps: sequencing the time sequence data of the historical meteorological data set to obtain a historical meteorological data sequence; extracting a block structure from the meteorological grid model, and performing block meteorological feature extraction operation on the historical meteorological data sequence according to the block structure to obtain a meteorological feature group sequence; performing time sequence feature extraction and residual feature mapping operation on the meteorological feature group sequence by using the meteorological grid model to obtain an analysis meteorological feature group sequence; performing sequence alignment on the meteorological feature group sequence by using the meteorological feature group sequence analysis to obtain an aligned meteorological feature group sequence; calculating a weather loss value of the weather analysis model according to the analysis weather feature group sequence and the alignment weather feature group sequence by using the following weather loss value algorithm:
Wherein W refers to the weather loss value, H, J refers to the sequence number, H refers to the feature total number of each of the analyzed weather feature groups in the analyzed weather feature group sequence, the feature total number of each of the analyzed weather feature groups in the analyzed weather feature group sequence is equal to the feature total number of each of the aligned weather feature groups in the aligned weather feature group sequence, J refers to the sequence length of the analyzed weather feature group sequence, the sequence length of the analyzed weather feature group sequence is equal to the sequence length of the analyzed weather feature group sequence, D j,h refers to the H analyzed weather feature in the J-th analyzed weather feature group in the analyzed weather feature group sequence, Y j,h refers to the H-th aligned weather feature in the J-th aligned weather feature group in the aligned weather feature group sequence, epsilon is a preset constant, lambda is a preset loss weight, and delta is a laplace operator symbol; carrying out iterative parameter updating on the meteorological grid model according to the meteorological loss value to obtain a meteorological analysis model;
performing joint training on the power grid structure model by utilizing the historical meteorological data set and the historical power grid data set to obtain a power grid analysis model;
And acquiring real-time power grid data and real-time meteorological data of the target area, calculating and analyzing the power grid data by using the meteorological analysis model, the power grid analysis model, the real-time power grid data and the real-time meteorological data, and performing power control on the target area according to the analyzed power grid data.
Optionally, the performing the grid structure analysis and the positioning modeling on the historical grid dataset to obtain a grid structure model includes:
Performing site detection on the historical power grid data set to obtain an electric website site set;
performing site coordinate positioning on the historical power grid data set according to the power grid site set to obtain a site position set;
Performing topological link analysis on the power grid site set according to the historical power grid data set to obtain a power grid connection structure;
and carrying out edge mapping on the site position set according to the power grid connection structure to obtain a power grid structure model.
Optionally, the performing geographic grid division and weather feature cluster mapping operation on the historical weather dataset to obtain a weather grid model includes:
extracting a historical meteorological map from the historical meteorological data set;
Sequentially carrying out geographic coordinate conversion and grid division on the historical meteorological map to obtain a meteorological block group;
performing weather feature mapping on the weather block group by using the historical weather data set to obtain a block weather feature group set;
Performing block clustering combination on the meteorological block groups according to the block meteorological feature group set to obtain a standard meteorological block group;
and performing grid mapping and model initialization operation on the standard meteorological block group to obtain a meteorological grid model.
Optionally, the block clustering combining is performed on the meteorological block group according to the block meteorological feature group set to obtain a standard meteorological block group, including:
Selecting weather blocks in the weather block group one by one as target weather blocks, and converging weather blocks adjacent to the target weather blocks in the weather block group into an adjacent block group;
The meteorological blocks in the adjacent block group are selected one by one to be used as target adjacent blocks, and the block distance between the target meteorological blocks and the target adjacent blocks is calculated by using the following block distance algorithm and the block meteorological feature group set:
Wherein, C is the block distance, μ is a preset distance countermeasure coefficient, Q is the total number of features of each block meteorological feature set in the block meteorological feature set, a q is the Q-th block meteorological feature in the block meteorological feature set corresponding to the target meteorological block, B q is the Q-th block meteorological feature in the block meteorological feature set corresponding to the target adjacent block, a x is the vector point multiplication symbol, a 5225 is the abscissa of the block midpoint of the target meteorological block, B x is the abscissa of the block midpoint of the target adjacent block, a y is the ordinate of the block midpoint of the target meteorological block, B y is the ordinate of the block midpoint of the target adjacent block, i is the absolute symbol, and i is the modulo symbol;
judging whether the block distance is smaller than a preset distance threshold value or not;
If not, returning to the step of selecting the meteorological blocks in the adjacent block group one by one as target adjacent blocks;
If yes, fusing the target meteorological blocks and the target adjacent blocks into target fusion blocks, updating the meteorological blocks in the meteorological block group by using the target fusion blocks, and returning to the step of selecting the meteorological blocks in the meteorological block group one by one as target meteorological blocks;
And taking the weather block group as a standard weather block group until the target weather block is the last weather block in the weather block group.
Optionally, the performing joint training on the grid structure model by using the historical meteorological data set and the historical grid data set to obtain a grid analysis model includes:
Extracting a site structure from the historical power grid data set to obtain a power grid distribution structure;
Performing time sequence sequencing and structure grouping operation on the historical meteorological data set by utilizing the power grid distribution structure to obtain a site meteorological data set sequence;
performing time sequence sorting and structure grouping operation on the historical power grid data set by utilizing the power grid distribution structure to obtain a historical power grid data set sequence;
Performing jump characteristic fusion operation on the site meteorological data group sequence and the historical power grid data group sequence to obtain a power grid meteorological characteristic group sequence;
And carrying out model training on the power grid structure model by using the power grid meteorological feature group sequence and the historical power grid data group sequence to obtain a power grid analysis model.
Optionally, the step of performing jump feature fusion operation on the site meteorological data group sequence and the historical grid data group sequence to obtain a grid meteorological feature group sequence includes:
extracting meteorological features of the site meteorological data group sequence to obtain a site meteorological feature group sequence;
Extracting power grid characteristics of the historical power grid data group sequence to obtain a historical power grid characteristic group sequence;
And carrying out feature fusion on the site meteorological feature group sequence and the historical power grid feature group sequence by using the following jump fusion algorithm to obtain a power grid meteorological feature group sequence:
Wherein Z i,j-1 refers to the ith grid meteorological feature in the J-1 th grid meteorological feature group in the grid meteorological feature group sequence, I refers to a sequence number, I refers to the total number of site meteorological features in each site meteorological feature group in the site meteorological feature group sequence, the total number of site meteorological features in each site meteorological feature group in the site meteorological feature group sequence is equal to the total number of historical grid features in each historical grid feature group in the historical grid feature group sequence, J refers to the sequence length of the site meteorological feature group sequence, the sequence length of the site meteorological feature group sequence is equal to the sequence length of the historical grid feature group sequence, softmax is a normalization function, Q i,j-1 is the ith historical grid feature in the J-1 th historical grid meteorological feature group in the site meteorological feature group sequence, R i,j is the ith site meteorological feature in the J site meteorological feature group in the site meteorological feature group sequence, α, β, γ is a preset attention coefficient matrix, and w () is a dimension transposed.
Optionally, the training the model of the power grid structure model by using the power grid meteorological feature group sequence and the historical power grid data group sequence to obtain a power grid analysis model includes:
Performing feature mapping operation on the power grid structure model by using the power grid meteorological feature group sequence to obtain an embedded power grid model;
performing recursive cyclic convolution and output mapping operation on the embedded power grid model to obtain an analysis power grid characteristic group sequence;
Performing sequence alignment on the historical power grid data set sequence by utilizing the analysis power grid characteristic set sequence to obtain an aligned power grid data set sequence;
Extracting power grid characteristics of the aligned power grid data group sequence to obtain an aligned power grid characteristic group sequence;
calculating a power grid loss value between the aligned power grid feature group sequence and the analysis power grid feature group sequence;
And carrying out iterative parameter updating on the embedded power grid model according to the power grid loss value to obtain a power grid analysis model.
Optionally, the calculating analysis grid data using the weather analysis model, the grid analysis model, the real-time grid data, and the real-time weather data includes:
Performing block weather feature extraction operation on the real-time weather data to obtain a real-time weather feature set;
performing weather analysis on the real-time weather feature set by using the weather analysis model to obtain a target weather feature set;
Performing structure grouping and power grid characteristic extraction operation on the real-time power grid data to obtain a real-time power grid characteristic group;
Performing jump characteristic fusion operation on the real-time power grid characteristic set and the target meteorological characteristic set to obtain a target power grid meteorological characteristic set;
calculating a target power grid characteristic group corresponding to the target power grid meteorological characteristic group by using the power grid analysis model;
And performing feature inverse mapping operation on the target power grid feature group to obtain analysis power grid data.
In order to solve the above problem, the present invention further provides a power control system based on source network load storage, the system comprising:
the structure modeling module is used for acquiring a historical power grid data set and a historical meteorological data set of a target area, and carrying out power grid structure analysis and positioning modeling on the historical power grid data set to obtain a power grid structure model;
The grid modeling module is used for carrying out geographic grid division and meteorological feature clustering mapping operation on the historical meteorological data set to obtain a meteorological grid model;
The weather training module is configured to perform time-series weather training on the weather grid model by using the historical weather data set to obtain a weather analysis model, where the time-series weather training on the weather grid model by using the historical weather data set to obtain the weather analysis model includes: sequencing the time sequence data of the historical meteorological data set to obtain a historical meteorological data sequence; extracting a block structure from the meteorological grid model, and performing block meteorological feature extraction operation on the historical meteorological data sequence according to the block structure to obtain a meteorological feature group sequence; performing time sequence feature extraction and residual feature mapping operation on the meteorological feature group sequence by using the meteorological grid model to obtain an analysis meteorological feature group sequence; performing sequence alignment on the meteorological feature group sequence by using the meteorological feature group sequence analysis to obtain an aligned meteorological feature group sequence; calculating a weather loss value of the weather analysis model according to the analysis weather feature group sequence and the alignment weather feature group sequence by using the following weather loss value algorithm:
Wherein W refers to the weather loss value, H, J refers to the sequence number, H refers to the feature total number of each of the analyzed weather feature groups in the analyzed weather feature group sequence, the feature total number of each of the analyzed weather feature groups in the analyzed weather feature group sequence is equal to the feature total number of each of the aligned weather feature groups in the aligned weather feature group sequence, J refers to the sequence length of the analyzed weather feature group sequence, the sequence length of the analyzed weather feature group sequence is equal to the sequence length of the analyzed weather feature group sequence, D j,h refers to the H analyzed weather feature in the J-th analyzed weather feature group in the analyzed weather feature group sequence, Y j,h refers to the H-th aligned weather feature in the J-th aligned weather feature group in the aligned weather feature group sequence, epsilon is a preset constant, lambda is a preset loss weight, and delta is a laplace operator symbol; carrying out iterative parameter updating on the meteorological grid model according to the meteorological loss value to obtain a meteorological analysis model;
The combined training module is used for carrying out combined training on the power grid structure model by utilizing the historical meteorological data set and the historical power grid data set to obtain a power grid analysis model;
And the power control module is used for acquiring real-time power grid data and real-time meteorological data of the target area, calculating analysis power grid data by using the meteorological analysis model, the power grid analysis model, the real-time power grid data and the real-time meteorological data, and controlling the power of the target area according to the analysis power grid data.
In order to solve the above-mentioned problems, the present invention also provides an electronic apparatus including:
At least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the source network charge storage based power control method described above.
According to the invention, through carrying out grid structure analysis and positioning modeling on the historical grid data set, a grid structure model is obtained, the complex relationship and dynamic change between grid sites can be more accurately simulated by utilizing the graph neural network, the accuracy of grid power analysis is improved, through generating the weather grid model according to the historical weather data set, the block simulation of weather data of a target area can be realized by utilizing the grid neural network, the spatial correlation of the weather data can be reserved, the generalization capability of the weather grid model is improved, and through carrying out time sequence weather training on the weather grid model by utilizing the historical weather data set, a weather analysis model is obtained, the weather data in the subsequent time period can be analyzed by combining the position relationship between areas and the time sequence relationship of the weather data, and the accuracy of weather data analysis is improved.
The power grid structure model is jointly trained by utilizing the historical meteorological data set and the historical power grid data set to obtain a power grid analysis model, the power grid data of a subsequent period of a target area can be analyzed and predicted by combining the real-time influence of the meteorological data on the power grid data and the periodicity rule of the power grid data, the accuracy of the power grid data analysis is improved, the power of each site of the power grid of the target area can be timely adjusted according to the analyzed power grid data of each site of the power grid of the target area in a future time period, the accurate adjustment of the power grid is realized, the supply and demand matching of electric energy is ensured, and the efficiency of the power control is further improved. Therefore, the power control method, the system and the equipment based on the source network charge storage can solve the problem of lower efficiency in power control.
Drawings
Fig. 1 is a flow chart of a power control method based on source network load storage according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of generating a power grid structure model according to an embodiment of the present invention;
FIG. 3 is a flow chart of generating a weather grid model according to an embodiment of the present invention;
Fig. 4 is a functional block diagram of a power control system based on source network load storage according to an embodiment of the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
The embodiment of the application provides a power control method based on source network charge storage. The execution main body of the power control method based on source network charge storage comprises at least one of electronic equipment, such as a server side, a terminal and the like, which can be configured to execute the method provided by the embodiment of the application. In other words, the power control method based on source network payload storage may be performed by software or hardware installed in a terminal device or a server device, where the software may be a blockchain platform. The service end includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like. The server may be an independent server, or may be a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms.
Referring to fig. 1, a flow chart of a power control method based on source network load storage according to an embodiment of the invention is shown. In this embodiment, the power control method based on source network load storage includes:
S1, acquiring a historical power grid data set and a historical meteorological data set of a target area, and carrying out power grid structure analysis and positioning modeling on the historical power grid data set to obtain a power grid structure model.
In the embodiment of the present invention, each historical grid data in the historical grid data set refers to a communication condition, power generation data, load data, energy storage data and the like of each node in the grid of the target area recorded in a past period of time, and each historical meteorological data in the historical meteorological data set refers to meteorological data of the target area recorded in a past period of time, such as temperature, humidity, wind speed, wind direction, weather and the like.
Specifically, the historical power grid data set and the historical meteorological data set can be obtained through a data crawling mode by a worker with corresponding authority, and the power grid structure model is a graph neural network model with a topological graph structure and composed of nodes and edges.
In the embodiment of the present invention, referring to fig. 2, the performing grid structure analysis and location modeling on the historical grid dataset to obtain a grid structure model includes:
S21, performing site detection on the historical power grid data set to obtain an electric website point set;
S22, positioning site coordinates of the historical power grid data set according to the power grid site set to obtain a site position set;
s23, carrying out topological link analysis on the power grid site set according to the historical power grid data set to obtain a power grid connection structure;
and S24, performing edge mapping on the site position set according to the power grid connection structure to obtain a power grid structure model.
In detail, the site detection refers to detecting the names of all power grid sites in the historical power grid data set by using a keyword matching method, and collecting the names into an electric website site set, wherein the power grid sites refer to important nodes for power generation, transmission, distribution and control in a power grid system, such as a wind power station, a power distribution station, a resident power substation, an industrial power substation, a power storage station and the like.
Specifically, the site coordinate positioning refers to matching position coordinate data corresponding to each power grid site in the power grid site set from the historical power grid data set, the topological link analysis refers to extracting connection relations of each site in the historical power grid data set, and a power grid connection structure is generated according to all the site connection relations by using a topological analysis method.
In detail, the edge mapping refers to mapping each site connection relation in the power grid connection structure between nodes corresponding to the site location set, and generating a power grid structure model of a graph structure.
Specifically, by carrying out grid structure analysis and positioning modeling on the historical grid data set, a grid structure model is obtained, complex relations and dynamic changes among grid sites can be simulated more accurately by using a graph neural network, and the accuracy of power analysis of the grid is improved.
And S2, performing geographic grid division and meteorological feature clustering mapping operation on the historical meteorological data set to obtain a meteorological grid model.
In the embodiment of the invention, the weather grid model refers to a time sequence grid neural network model for weather prediction, the input of the weather grid model is the weather characteristics of each grid area of the target area in the previous time period, and the output of the weather grid model is the weather characteristics of each grid area of the target area in the next time period.
In the embodiment of the present invention, referring to fig. 3, the performing geographic grid division and weather feature cluster mapping operation on the historical weather dataset to obtain a weather grid model includes:
S31, extracting a historical meteorological map from the historical meteorological data set;
S32, sequentially performing geographic coordinate conversion and grid division on the historical meteorological map to obtain a meteorological block group;
S33, performing weather feature mapping on the weather block set by using the historical weather data set to obtain a block weather feature set;
s34, carrying out block clustering combination on the meteorological block groups according to the block meteorological feature group set to obtain a standard meteorological block group;
And S35, performing grid mapping and model initialization operation on the standard meteorological block group to obtain a meteorological grid model.
In detail, the historical weather map refers to a weather map of the target area, the geographic coordinate transformation refers to transforming the coordinates of the historical weather map according to the actual size scale and the area position coordinates of the target area, the grid division refers to dividing the historical weather map after coordinate transformation into a plurality of weather blocks with equal size according to a preset grid size, and all the weather blocks are collected into a weather block group.
Specifically, the step of mapping the meteorological features of the meteorological block group by using the historical meteorological data set to obtain a block meteorological feature group set refers to selecting meteorological blocks in the meteorological block group one by one as target meteorological blocks, screening out a block meteorological data set corresponding to the target meteorological blocks from the historical meteorological data set, vectorizing the block meteorological data set into a block meteorological feature set, and collecting all the block meteorological feature sets into a block meteorological feature group set.
In detail, the block clustering merging is performed on the meteorological block group according to the block meteorological feature group set to obtain a standard meteorological block group, which includes:
Selecting weather blocks in the weather block group one by one as target weather blocks, and converging weather blocks adjacent to the target weather blocks in the weather block group into an adjacent block group;
The meteorological blocks in the adjacent block group are selected one by one to be used as target adjacent blocks, and the block distance between the target meteorological blocks and the target adjacent blocks is calculated by using the following block distance algorithm and the block meteorological feature group set:
Wherein, C is the block distance, μ is a preset distance countermeasure coefficient, Q is the total number of features of each block meteorological feature set in the block meteorological feature set, a q is the Q-th block meteorological feature in the block meteorological feature set corresponding to the target meteorological block, B q is the Q-th block meteorological feature in the block meteorological feature set corresponding to the target adjacent block, a x is the vector point multiplication symbol, a 5225 is the abscissa of the block midpoint of the target meteorological block, B x is the abscissa of the block midpoint of the target adjacent block, a y is the ordinate of the block midpoint of the target meteorological block, B y is the ordinate of the block midpoint of the target adjacent block, i is the absolute symbol, and i is the modulo symbol;
judging whether the block distance is smaller than a preset distance threshold value or not;
If not, returning to the step of selecting the meteorological blocks in the adjacent block group one by one as target adjacent blocks;
If yes, fusing the target meteorological blocks and the target adjacent blocks into target fusion blocks, updating the meteorological blocks in the meteorological block group by using the target fusion blocks, and returning to the step of selecting the meteorological blocks in the meteorological block group one by one as target meteorological blocks;
And taking the weather block group as a standard weather block group until the target weather block is the last weather block in the weather block group.
In detail, the block distance algorithm can combine the difference of meteorological features and the distance difference between blocks to realize the fusion of blocks with similar meteorological, thereby reducing the grid complexity of a meteorological grid model and reserving the spatial relevance of meteorological data.
Specifically, the grid mapping refers to extracting a block structure from the standard meteorological block group, performing grid mapping on the block structure to obtain an initial grid model, and initializing a time sequence neural structure layer for the initial grid model to obtain a meteorological grid model.
According to the embodiment of the invention, the weather grid model is generated according to the historical weather data set, so that the grid neural network can be utilized to realize the block simulation of the weather data of the target area, the spatial correlation of the weather data can be reserved, and the generalization capability of the weather grid model is improved.
S3, performing time sequence weather training on the weather grid model by using the historical weather data set to obtain a weather analysis model.
In the embodiment of the invention, the weather analysis model is a trained weather grid model, and the weather analysis model can predict weather features of each grid region of the target region in the next time period according to weather features of each grid region of the target region in the previous time period.
In the embodiment of the present invention, the time-series weather training is performed on the weather grid model by using the historical weather data set to obtain a weather analysis model, including:
sequencing the time sequence data of the historical meteorological data set to obtain a historical meteorological data sequence;
extracting a block structure from the meteorological grid model, and performing block meteorological feature extraction operation on the historical meteorological data sequence according to the block structure to obtain a meteorological feature group sequence;
performing time sequence feature extraction and residual feature mapping operation on the meteorological feature group sequence by using the meteorological grid model to obtain an analysis meteorological feature group sequence;
performing sequence alignment on the meteorological feature group sequence by using the meteorological feature group sequence analysis to obtain an aligned meteorological feature group sequence;
Calculating a weather loss value of the weather analysis model according to the analysis weather feature group sequence and the alignment weather feature group sequence by using the following weather loss value algorithm:
Wherein W refers to the weather loss value, H, J refers to the sequence number, H refers to the feature total number of each of the analyzed weather feature groups in the analyzed weather feature group sequence, the feature total number of each of the analyzed weather feature groups in the analyzed weather feature group sequence is equal to the feature total number of each of the aligned weather feature groups in the aligned weather feature group sequence, J refers to the sequence length of the analyzed weather feature group sequence, the sequence length of the analyzed weather feature group sequence is equal to the sequence length of the analyzed weather feature group sequence, D j,h refers to the H analyzed weather feature in the J-th analyzed weather feature group in the analyzed weather feature group sequence, Y j,h refers to the H-th aligned weather feature in the J-th aligned weather feature group in the aligned weather feature group sequence, epsilon is a preset constant, lambda is a preset loss weight, and delta is a laplace operator symbol;
and carrying out iterative parameter updating on the meteorological grid model according to the meteorological loss value to obtain a meteorological analysis model.
In detail, the time sequence data sorting means that the historical meteorological data sets are sorted according to a fixed time period length and a time sequence, and historical meteorological data in each period length are collected into a historical meteorological data sequence.
Specifically, the block structure refers to a composition structure of each standard weather block in a standard weather block group corresponding to the weather grid model, the time sequence feature extraction refers to time dependent feature extraction of the weather feature group sequence by using a gating structure in the weather grid model, and the residual feature mapping refers to mapping of time dependent features by using a full connection layer and a residual connection layer in the weather grid model.
Specifically, the meteorological loss value algorithm can determine the meteorological loss value of the model according to the difference of the meteorological features of each corresponding block in the target area in the corresponding time period according to the analysis meteorological feature group sequence and the alignment meteorological feature group sequence, so that the accuracy of model training is improved.
In detail, the sequence alignment refers to assembling partial meteorological feature groups with the same timestamp as the analysis meteorological feature group sequence in the meteorological feature group sequence into an aligned meteorological feature group sequence, and the iterative parameter updating method can be a gradient descent algorithm.
According to the embodiment of the invention, the time sequence weather training is carried out on the weather grid model by utilizing the historical weather data set to obtain the weather analysis model, so that the weather data in the subsequent time period can be analyzed by combining the position relation between areas and the time sequence relation of the weather data, and the accuracy of the weather data analysis is improved.
And S4, performing joint training on the power grid structure model by using the historical meteorological data set and the historical power grid data set to obtain a power grid analysis model.
In the embodiment of the invention, the power grid analysis model is a graph cycle neural network model which is input as the power grid characteristics of the previous time period and the meteorological characteristics of the next time period and output as the power grid characteristics of the next time period.
In detail, since the power grid data are more influenced by the use habit, and the power consumption peak period and the power generation data are more influenced by the weather data such as the temperature, the solar irradiation intensity, the wind power, the wind direction and the like, the power grid structure model can be jointly trained according to the historical weather data set and the historical power grid data set, and a power grid analysis model is obtained.
In the embodiment of the present invention, the performing, by using the historical meteorological data set and the historical grid data set, the joint training on the grid structure model to obtain a grid analysis model includes:
Extracting a site structure from the historical power grid data set to obtain a power grid distribution structure;
Performing time sequence sequencing and structure grouping operation on the historical meteorological data set by utilizing the power grid distribution structure to obtain a site meteorological data set sequence;
performing time sequence sorting and structure grouping operation on the historical power grid data set by utilizing the power grid distribution structure to obtain a historical power grid data set sequence;
Performing jump characteristic fusion operation on the site meteorological data group sequence and the historical power grid data group sequence to obtain a power grid meteorological characteristic group sequence;
And carrying out model training on the power grid structure model by using the power grid meteorological feature group sequence and the historical power grid data group sequence to obtain a power grid analysis model.
In detail, the geographical distribution structure refers to a distribution position and a distribution structure of each station in the target area corresponding to the historical grid dataset, and the station structure extraction can be performed by using a keyword retrieval method.
Specifically, the structure grouping refers to grouping the historical weather data sets and the historical grid data sets after sequencing according to interval positions corresponding to all grid sites in the grid distribution structure, each site weather data set in the site weather data set sequence corresponds to weather data of each site area in the target area in a time period, each site weather data in the site weather data set corresponds to weather data of one site area, each historical grid data set in the historical grid data set sequence corresponds to grid data of each site area in the target area in a time period, and each historical grid data in the historical grid data set corresponds to grid data of one site area.
Specifically, the step of performing jump feature fusion operation on the site meteorological data group sequence and the historical power grid data group sequence to obtain a power grid meteorological feature group sequence includes:
extracting meteorological features of the site meteorological data group sequence to obtain a site meteorological feature group sequence;
Extracting power grid characteristics of the historical power grid data group sequence to obtain a historical power grid characteristic group sequence;
And carrying out feature fusion on the site meteorological feature group sequence and the historical power grid feature group sequence by using the following jump fusion algorithm to obtain a power grid meteorological feature group sequence:
Wherein Z i,j-1 refers to the ith grid meteorological feature in the J-1 th grid meteorological feature group in the grid meteorological feature group sequence, I refers to a sequence number, I refers to the total number of site meteorological features in each site meteorological feature group in the site meteorological feature group sequence, the total number of site meteorological features in each site meteorological feature group in the site meteorological feature group sequence is equal to the total number of historical grid features in each historical grid feature group in the historical grid feature group sequence, J refers to the sequence length of the site meteorological feature group sequence, the sequence length of the site meteorological feature group sequence is equal to the sequence length of the historical grid feature group sequence, softmax is a normalization function, Q i,j-1 is the ith historical grid feature in the J-1 th historical grid meteorological feature group in the site meteorological feature group sequence, R i,j is the ith site meteorological feature in the J site meteorological feature group in the site meteorological feature group sequence, α, β, γ is a preset attention coefficient matrix, and w () is a dimension transposed.
Specifically, the jump fusion algorithm can be combined with the attention feature fusion method to realize fusion of the site meteorological feature group sequence and the historical power grid feature group sequence, and meanwhile, the time period difference between the power grid features and the meteorological features during fusion is ensured, so that the characterization degree of the power grid meteorological features obtained after fusion is improved.
In detail, the training the model of the power grid structure model by using the power grid meteorological feature group sequence and the historical power grid data group sequence to obtain a power grid analysis model comprises the following steps:
Performing feature mapping operation on the power grid structure model by using the power grid meteorological feature group sequence to obtain an embedded power grid model;
performing recursive cyclic convolution and output mapping operation on the embedded power grid model to obtain an analysis power grid characteristic group sequence;
Performing sequence alignment on the historical power grid data set sequence by utilizing the analysis power grid characteristic set sequence to obtain an aligned power grid data set sequence;
Extracting power grid characteristics of the aligned power grid data group sequence to obtain an aligned power grid characteristic group sequence;
calculating a power grid loss value between the aligned power grid feature group sequence and the analysis power grid feature group sequence;
And carrying out iterative parameter updating on the embedded power grid model according to the power grid loss value to obtain a power grid analysis model.
In detail, the step of performing feature embedding operation on the power grid structure model by using the power grid meteorological feature group sequence to obtain an embedded power grid model refers to embedding each power grid meteorological feature group in the power grid meteorological feature group sequence into a model node and a model edge of the power grid structure model according to a corresponding relation between a time sequence and the power grid distribution structure.
In detail, a recurrent neural network or a long and short term memory network can be utilized to carry out recurrent convolution, so that the representation of the nodes is recursively calculated or iteratively updated to capture the connection relation and the time sequence between the nodes, the output mapping can be realized through a full connection layer, and the power grid loss value can be calculated by utilizing a mean square error loss value algorithm.
In the embodiment of the invention, the historical meteorological data set and the historical power grid data set are utilized to carry out joint training on the power grid structure model to obtain the power grid analysis model, so that the real-time influence of the meteorological data on the power grid data and the periodic rule of the power grid data can be combined to carry out analysis and prediction on the power grid data in the subsequent period of the target area, and the accuracy of the power grid data analysis is improved.
And S5, acquiring real-time power grid data and real-time meteorological data of the target area, calculating analysis power grid data by using the meteorological analysis model, the power grid analysis model, the real-time power grid data and the real-time meteorological data, and performing power control on the target area according to the analysis power grid data.
In the embodiment of the invention, the real-time power grid data refer to power grid data corresponding to each site in the target area in the current time period, and the real-time meteorological data refer to meteorological data of the target area in the current time period.
In detail, the calculating and analyzing the power grid data by using the weather analysis model, the power grid analysis model, the real-time power grid data and the real-time weather data includes:
Performing block weather feature extraction operation on the real-time weather data to obtain a real-time weather feature set;
performing weather analysis on the real-time weather feature set by using the weather analysis model to obtain a target weather feature set;
Performing structure grouping and power grid characteristic extraction operation on the real-time power grid data to obtain a real-time power grid characteristic group;
Performing jump characteristic fusion operation on the real-time power grid characteristic set and the target meteorological characteristic set to obtain a target power grid meteorological characteristic set;
calculating a target power grid characteristic group corresponding to the target power grid meteorological characteristic group by using the power grid analysis model;
And performing feature inverse mapping operation on the target power grid feature group to obtain analysis power grid data.
In detail, the block weather feature extraction operation is the same as the block weather feature extraction in the step S3, and the feature inverse mapping operation refers to an inverse operation of the grid feature extraction operation.
Specifically, the step of performing power control on the target area according to the analysis power grid data refers to timely adjusting the power generation power and the charging and discharging power of the corresponding stations according to the power grid data such as the communication condition, the power generation data, the load data, the energy storage data and the like of each station obtained by analysis in the analysis power grid data, so that power balance of the power grid in a subsequent time period is achieved, and supply and demand matching of electric energy is ensured.
In the embodiment of the invention, the power of each power grid site of the target area can be timely adjusted according to the analyzed power grid data of each power grid site of the target area in a future time period by controlling the power of the target area according to the analyzed power grid data, so that the accurate adjustment of the power grid is realized, the supply and demand matching of electric energy is ensured, and the efficiency of power control is further improved.
According to the invention, through carrying out grid structure analysis and positioning modeling on the historical grid data set, a grid structure model is obtained, the complex relationship and dynamic change between grid sites can be more accurately simulated by utilizing the graph neural network, the accuracy of grid power analysis is improved, through generating the weather grid model according to the historical weather data set, the block simulation of weather data of a target area can be realized by utilizing the grid neural network, the spatial correlation of the weather data can be reserved, the generalization capability of the weather grid model is improved, and through carrying out time sequence weather training on the weather grid model by utilizing the historical weather data set, a weather analysis model is obtained, the weather data in the subsequent time period can be analyzed by combining the position relationship between areas and the time sequence relationship of the weather data, and the accuracy of weather data analysis is improved.
The power grid structure model is jointly trained by utilizing the historical meteorological data set and the historical power grid data set to obtain a power grid analysis model, the power grid data of a subsequent period of a target area can be analyzed and predicted by combining the real-time influence of the meteorological data on the power grid data and the periodicity rule of the power grid data, the accuracy of the power grid data analysis is improved, the power of each site of the power grid of the target area can be timely adjusted according to the analyzed power grid data of each site of the power grid of the target area in a future time period, the accurate adjustment of the power grid is realized, the supply and demand matching of electric energy is ensured, and the efficiency of the power control is further improved. Therefore, the power control method based on the source network charge storage can solve the problem of lower efficiency in power control.
Fig. 4 is a functional block diagram of a power control system based on source network load storage according to an embodiment of the present invention.
The power control system 100 based on source network charge storage according to the present invention may be installed in an electronic device. Depending on the functions implemented, the source network charge storage based power control system 100 may include a structural modeling module 101, a grid modeling module 102, a weather training module 103, a joint training module 104, and a power control module 105. The module of the invention, which may also be referred to as a unit, refers to a series of computer program segments, which are stored in the memory of the electronic device, capable of being executed by the processor of the electronic device and of performing a fixed function.
In the present embodiment, the functions concerning the respective modules/units are as follows:
The structure modeling module 101 is configured to obtain a historical power grid dataset and a historical meteorological dataset of a target area, perform power grid structure analysis and positioning modeling on the historical power grid dataset, and obtain a power grid structure model;
the grid modeling module 102 is configured to perform geographic grid division and weather feature cluster mapping operation on the historical weather dataset to obtain a weather grid model;
The weather training module 103 is configured to perform time-series weather training on the weather grid model by using the historical weather data set to obtain a weather analysis model, where the performing time-series weather training on the weather grid model by using the historical weather data set to obtain the weather analysis model includes: sequencing the time sequence data of the historical meteorological data set to obtain a historical meteorological data sequence; extracting a block structure from the meteorological grid model, and performing block meteorological feature extraction operation on the historical meteorological data sequence according to the block structure to obtain a meteorological feature group sequence; performing time sequence feature extraction and residual feature mapping operation on the meteorological feature group sequence by using the meteorological grid model to obtain an analysis meteorological feature group sequence; performing sequence alignment on the meteorological feature group sequence by using the meteorological feature group sequence analysis to obtain an aligned meteorological feature group sequence; calculating a weather loss value of the weather analysis model according to the analysis weather feature group sequence and the alignment weather feature group sequence by using the following weather loss value algorithm:
Wherein W refers to the weather loss value, H, J refers to the sequence number, H refers to the feature total number of each of the analyzed weather feature groups in the analyzed weather feature group sequence, the feature total number of each of the analyzed weather feature groups in the analyzed weather feature group sequence is equal to the feature total number of each of the aligned weather feature groups in the aligned weather feature group sequence, J refers to the sequence length of the analyzed weather feature group sequence, the sequence length of the analyzed weather feature group sequence is equal to the sequence length of the analyzed weather feature group sequence, D j,h refers to the H analyzed weather feature in the J-th analyzed weather feature group in the analyzed weather feature group sequence, Y j,h refers to the H-th aligned weather feature in the J-th aligned weather feature group in the aligned weather feature group sequence, epsilon is a preset constant, lambda is a preset loss weight, and delta is a laplace operator symbol; carrying out iterative parameter updating on the meteorological grid model according to the meteorological loss value to obtain a meteorological analysis model;
The joint training module 104 is configured to perform joint training on the power grid structure model by using the historical meteorological data set and the historical power grid data set to obtain a power grid analysis model;
The power control module 105 is configured to obtain real-time power grid data and real-time weather data of the target area, calculate analysis power grid data using the weather analysis model, the power grid analysis model, the real-time power grid data and the real-time weather data, and perform power control on the target area according to the analysis power grid data.
In detail, each module in the power control system 100 based on source network load storage in the embodiment of the present invention adopts the same technical means as the power control method based on source network load storage described in fig. 1 to 3, and can generate the same technical effects, which is not described herein.
The embodiment of the invention provides a structural schematic diagram of electronic equipment for realizing a power control method based on source network charge storage.
The electronic device (not shown) may include a processor, a memory, a communication bus, and a communication interface, and may also include a computer program stored in the memory and executable on the processor, such as a power control program stored on a source network load basis.
The processor may be formed by an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or may be formed by a plurality of integrated circuits packaged with the same function or different functions, including one or more central processing units (Central Processing unit, CPU), microprocessors, digital processing chips, graphics processors, and combinations of various control chips. The processor is a Control Unit (Control Unit) of the electronic device, connects various components of the entire electronic device using various interfaces and lines, and executes various functions of the electronic device and processes data by running or executing programs or modules stored in the memory (e.g., executing a power Control program stored based on source network load, etc.), and calling data stored in the memory.
The memory includes at least one type of readable storage medium including flash memory, removable hard disk, multimedia card, card memory (e.g., SD or DX memory, etc.), magnetic memory, magnetic disk, optical disk, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device, such as a mobile hard disk of the electronic device. The memory 11 may also be an external storage device of the electronic device in other embodiments, such as a plug-in mobile hard disk, a smart memory card (SMART MEDIA CARD, SMC), a Secure Digital (SD) card, a flash memory card (FLASH CARD) or the like, which are provided on the electronic device. Further, the memory may also include both internal storage units and external storage devices of the electronic device. The memory may be used not only to store application software installed in an electronic device and various types of data, such as codes of a power control program stored based on a source network load, but also to temporarily store data that has been output or is to be output.
The communication bus may be a peripheral component interconnect standard (PERIPHERAL COMPONENT INTERCONNECT, PCI) bus or an extended industry standard architecture (extended industry standard architecture, EISA) bus, etc. The bus may be classified as an address bus, a data bus, a control bus, etc. The bus is arranged to enable a connection communication between the memory and at least one processor or the like.
The communication interface is used for communication between the electronic equipment and other equipment, and comprises a network interface and a user interface. Optionally, the network interface may include a wired interface and/or a wireless interface (e.g., WI-FI interface, bluetooth interface, etc.), typically used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a Display (Display), an input unit such as a Keyboard (Keyboard), or alternatively a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like. Among other things, the display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device and for displaying a visual user interface.
Only an electronic device having components is shown, and it will be understood by those skilled in the art that the structures shown in the figures do not limit the electronic device, and may include fewer or more components than shown, or may combine certain components, or a different arrangement of components.
For example, although not shown, the electronic device may further include a power source (such as a battery) for supplying power to the respective components, and preferably, the power source may be logically connected to the at least one processor through a power management device, so that functions of charge management, discharge management, power consumption management, and the like are implemented through the power management device. The power supply may also include one or more of any of a direct current or alternating current power supply, recharging device, power failure detection circuit, power converter or inverter, power status indicator, etc. The electronic device may further include various sensors, bluetooth modules, wi-Fi modules, etc., which are not described herein.
It should be understood that the embodiments described are for illustrative purposes only and are not limited to this configuration in the scope of the patent application.
In the several embodiments provided in the present invention, it should be understood that the disclosed apparatus, system and method may be implemented in other manners. For example, the system embodiments described above are merely illustrative, e.g., the division of the modules is merely a logical function division, and other manners of division may be implemented in practice.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
The embodiment of the application can acquire and process the related data based on the artificial intelligence technology. Wherein artificial intelligence (ARTIFICIAL INTELLIGENCE, AI) is the theory, method, technique, and application system that uses a digital computer or a digital computer-controlled machine to simulate, extend, and expand human intelligence, sense the environment, acquire knowledge, and use knowledge to obtain optimal results.
Furthermore, it is evident that the word "comprising" does not exclude other elements or steps, and that the singular does not exclude a plurality. Multiple units or systems set forth in the system embodiments may also be implemented by one unit or system in software or hardware. The terms first, second, etc. are used to denote a name, but not any particular order.
Finally, it should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.
Claims (10)
1. A power control method based on source network charge storage, the method comprising:
Acquiring a historical power grid data set and a historical meteorological data set of a target area, and carrying out power grid structure analysis and positioning modeling on the historical power grid data set to obtain a power grid structure model;
performing geographic grid division and meteorological feature clustering mapping operation on the historical meteorological data set to obtain a meteorological grid model;
Performing time sequence weather training on the weather grid model by using the historical weather data set to obtain a weather analysis model, wherein the performing time sequence weather training on the weather grid model by using the historical weather data set to obtain the weather analysis model comprises the following steps: sequencing the time sequence data of the historical meteorological data set to obtain a historical meteorological data sequence; extracting a block structure from the meteorological grid model, and performing block meteorological feature extraction operation on the historical meteorological data sequence according to the block structure to obtain a meteorological feature group sequence; performing time sequence feature extraction and residual feature mapping operation on the meteorological feature group sequence by using the meteorological grid model to obtain an analysis meteorological feature group sequence; performing sequence alignment on the meteorological feature group sequence by using the meteorological feature group sequence analysis to obtain an aligned meteorological feature group sequence; calculating a weather loss value of the weather analysis model according to the analysis weather feature group sequence and the alignment weather feature group sequence by using the following weather loss value algorithm:
Wherein W refers to the weather loss value, H, J refers to the sequence number, H refers to the feature total number of each of the analyzed weather feature groups in the analyzed weather feature group sequence, the feature total number of each of the analyzed weather feature groups in the analyzed weather feature group sequence is equal to the feature total number of each of the aligned weather feature groups in the aligned weather feature group sequence, J refers to the sequence length of the analyzed weather feature group sequence, the sequence length of the analyzed weather feature group sequence is equal to the sequence length of the analyzed weather feature group sequence, D j,h refers to the H analyzed weather feature in the J-th analyzed weather feature group in the analyzed weather feature group sequence, Y j,h refers to the H-th aligned weather feature in the J-th aligned weather feature group in the aligned weather feature group sequence, epsilon is a preset constant, lambda is a preset loss weight, and delta is a laplace operator symbol; carrying out iterative parameter updating on the meteorological grid model according to the meteorological loss value to obtain a meteorological analysis model;
performing joint training on the power grid structure model by utilizing the historical meteorological data set and the historical power grid data set to obtain a power grid analysis model;
And acquiring real-time power grid data and real-time meteorological data of the target area, calculating and analyzing the power grid data by using the meteorological analysis model, the power grid analysis model, the real-time power grid data and the real-time meteorological data, and performing power control on the target area according to the analyzed power grid data.
2. The power control method based on source network load storage according to claim 1, wherein the performing a network structure analysis and a positioning modeling on the historical network data set to obtain a network structure model includes:
Performing site detection on the historical power grid data set to obtain an electric website site set;
performing site coordinate positioning on the historical power grid data set according to the power grid site set to obtain a site position set;
Performing topological link analysis on the power grid site set according to the historical power grid data set to obtain a power grid connection structure;
and carrying out edge mapping on the site position set according to the power grid connection structure to obtain a power grid structure model.
3. The power control method based on source network load storage according to claim 1, wherein said performing geographic meshing and weather feature cluster mapping operations on the historical weather dataset to obtain a weather mesh model comprises:
extracting a historical meteorological map from the historical meteorological data set;
Sequentially carrying out geographic coordinate conversion and grid division on the historical meteorological map to obtain a meteorological block group;
performing weather feature mapping on the weather block group by using the historical weather data set to obtain a block weather feature group set;
Performing block clustering combination on the meteorological block groups according to the block meteorological feature group set to obtain a standard meteorological block group;
and performing grid mapping and model initialization operation on the standard meteorological block group to obtain a meteorological grid model.
4. The method for power control based on source network load storage according to claim 3, wherein said performing block cluster merging on said meteorological block group according to said block meteorological feature group set to obtain a standard meteorological block group comprises:
Selecting weather blocks in the weather block group one by one as target weather blocks, and converging weather blocks adjacent to the target weather blocks in the weather block group into an adjacent block group;
The meteorological blocks in the adjacent block group are selected one by one to be used as target adjacent blocks, and the block distance between the target meteorological blocks and the target adjacent blocks is calculated by using the following block distance algorithm and the block meteorological feature group set:
Wherein, C is the block distance, μ is a preset distance countermeasure coefficient, Q is the total number of features of each block meteorological feature set in the block meteorological feature set, a q is the Q-th block meteorological feature in the block meteorological feature set corresponding to the target meteorological block, B q is the Q-th block meteorological feature in the block meteorological feature set corresponding to the target adjacent block, a x is the vector point multiplication symbol, a 5225 is the abscissa of the block midpoint of the target meteorological block, B x is the abscissa of the block midpoint of the target adjacent block, a y is the ordinate of the block midpoint of the target meteorological block, B y is the ordinate of the block midpoint of the target adjacent block, i is the absolute symbol, and i is the modulo symbol;
judging whether the block distance is smaller than a preset distance threshold value or not;
If not, returning to the step of selecting the meteorological blocks in the adjacent block group one by one as target adjacent blocks;
If yes, fusing the target meteorological blocks and the target adjacent blocks into target fusion blocks, updating the meteorological blocks in the meteorological block group by using the target fusion blocks, and returning to the step of selecting the meteorological blocks in the meteorological block group one by one as target meteorological blocks;
And taking the weather block group as a standard weather block group until the target weather block is the last weather block in the weather block group.
5. The power control method based on source network load storage according to claim 1, wherein the performing joint training on the grid structure model by using the historical meteorological data set and the historical grid data set to obtain a grid analysis model comprises:
Extracting a site structure from the historical power grid data set to obtain a power grid distribution structure;
Performing time sequence sequencing and structure grouping operation on the historical meteorological data set by utilizing the power grid distribution structure to obtain a site meteorological data set sequence;
performing time sequence sorting and structure grouping operation on the historical power grid data set by utilizing the power grid distribution structure to obtain a historical power grid data set sequence;
Performing jump characteristic fusion operation on the site meteorological data group sequence and the historical power grid data group sequence to obtain a power grid meteorological characteristic group sequence;
And carrying out model training on the power grid structure model by using the power grid meteorological feature group sequence and the historical power grid data group sequence to obtain a power grid analysis model.
6. The power control method based on source network load storage according to claim 5, wherein the step of performing jump feature fusion operation on the site meteorological data group sequence and the historical power grid data group sequence to obtain a power grid meteorological feature group sequence includes:
extracting meteorological features of the site meteorological data group sequence to obtain a site meteorological feature group sequence;
Extracting power grid characteristics of the historical power grid data group sequence to obtain a historical power grid characteristic group sequence;
And carrying out feature fusion on the site meteorological feature group sequence and the historical power grid feature group sequence by using the following jump fusion algorithm to obtain a power grid meteorological feature group sequence:
Wherein Z i,j-1 refers to the ith grid meteorological feature in the J-1 th grid meteorological feature group in the grid meteorological feature group sequence, I refers to a sequence number, I refers to the total number of site meteorological features in each site meteorological feature group in the site meteorological feature group sequence, the total number of site meteorological features in each site meteorological feature group in the site meteorological feature group sequence is equal to the total number of historical grid features in each historical grid feature group in the historical grid feature group sequence, J refers to the sequence length of the site meteorological feature group sequence, the sequence length of the site meteorological feature group sequence is equal to the sequence length of the historical grid feature group sequence, softmax is a normalization function, Q i,j-1 is the ith historical grid feature in the J-1 th historical grid meteorological feature group in the site meteorological feature group sequence, R i,j is the ith site meteorological feature in the J site meteorological feature group in the site meteorological feature group sequence, α, β, γ is a preset attention coefficient matrix, and w () is a dimension transposed.
7. The power control method based on source network load storage according to claim 5, wherein said model training said grid structure model using said grid meteorological feature group sequence and said historical grid data group sequence to obtain a grid analysis model, comprises:
Performing feature mapping operation on the power grid structure model by using the power grid meteorological feature group sequence to obtain an embedded power grid model;
performing recursive cyclic convolution and output mapping operation on the embedded power grid model to obtain an analysis power grid characteristic group sequence;
Performing sequence alignment on the historical power grid data set sequence by utilizing the analysis power grid characteristic set sequence to obtain an aligned power grid data set sequence;
Extracting power grid characteristics of the aligned power grid data group sequence to obtain an aligned power grid characteristic group sequence;
calculating a power grid loss value between the aligned power grid feature group sequence and the analysis power grid feature group sequence;
And carrying out iterative parameter updating on the embedded power grid model according to the power grid loss value to obtain a power grid analysis model.
8. The source network charge storage-based power control method of claim 1, wherein said calculating analysis grid data using said weather analysis model, said grid analysis model, said real-time grid data, and said real-time weather data comprises:
Performing block weather feature extraction operation on the real-time weather data to obtain a real-time weather feature set;
performing weather analysis on the real-time weather feature set by using the weather analysis model to obtain a target weather feature set;
Performing structure grouping and power grid characteristic extraction operation on the real-time power grid data to obtain a real-time power grid characteristic group;
Performing jump characteristic fusion operation on the real-time power grid characteristic set and the target meteorological characteristic set to obtain a target power grid meteorological characteristic set;
calculating a target power grid characteristic group corresponding to the target power grid meteorological characteristic group by using the power grid analysis model;
And performing feature inverse mapping operation on the target power grid feature group to obtain analysis power grid data.
9. A power control system based on source network charge storage, the system comprising:
the structure modeling module is used for acquiring a historical power grid data set and a historical meteorological data set of a target area, and carrying out power grid structure analysis and positioning modeling on the historical power grid data set to obtain a power grid structure model;
The grid modeling module is used for carrying out geographic grid division and meteorological feature clustering mapping operation on the historical meteorological data set to obtain a meteorological grid model;
The weather training module is configured to perform time-series weather training on the weather grid model by using the historical weather data set to obtain a weather analysis model, where the time-series weather training on the weather grid model by using the historical weather data set to obtain the weather analysis model includes: sequencing the time sequence data of the historical meteorological data set to obtain a historical meteorological data sequence; extracting a block structure from the meteorological grid model, and performing block meteorological feature extraction operation on the historical meteorological data sequence according to the block structure to obtain a meteorological feature group sequence; performing time sequence feature extraction and residual feature mapping operation on the meteorological feature group sequence by using the meteorological grid model to obtain an analysis meteorological feature group sequence; performing sequence alignment on the meteorological feature group sequence by using the meteorological feature group sequence analysis to obtain an aligned meteorological feature group sequence; calculating a weather loss value of the weather analysis model according to the analysis weather feature group sequence and the alignment weather feature group sequence by using the following weather loss value algorithm:
Wherein W refers to the weather loss value, H, J refers to the sequence number, H refers to the feature total number of each of the analyzed weather feature groups in the analyzed weather feature group sequence, the feature total number of each of the analyzed weather feature groups in the analyzed weather feature group sequence is equal to the feature total number of each of the aligned weather feature groups in the aligned weather feature group sequence, J refers to the sequence length of the analyzed weather feature group sequence, the sequence length of the analyzed weather feature group sequence is equal to the sequence length of the analyzed weather feature group sequence, D j,h refers to the H analyzed weather feature in the J-th analyzed weather feature group in the analyzed weather feature group sequence, Y j,h refers to the H-th aligned weather feature in the J-th aligned weather feature group in the aligned weather feature group sequence, epsilon is a preset constant, lambda is a preset loss weight, and delta is a laplace operator symbol; carrying out iterative parameter updating on the meteorological grid model according to the meteorological loss value to obtain a meteorological analysis model;
The combined training module is used for carrying out combined training on the power grid structure model by utilizing the historical meteorological data set and the historical power grid data set to obtain a power grid analysis model;
And the power control module is used for acquiring real-time power grid data and real-time meteorological data of the target area, calculating analysis power grid data by using the meteorological analysis model, the power grid analysis model, the real-time power grid data and the real-time meteorological data, and controlling the power of the target area according to the analysis power grid data.
10. An electronic device, the electronic device comprising:
At least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the source network charge storage based power control method of any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410311320.2A CN118336829A (en) | 2024-03-19 | 2024-03-19 | Power control method, system and equipment based on source network charge storage |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410311320.2A CN118336829A (en) | 2024-03-19 | 2024-03-19 | Power control method, system and equipment based on source network charge storage |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118336829A true CN118336829A (en) | 2024-07-12 |
Family
ID=91775410
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410311320.2A Pending CN118336829A (en) | 2024-03-19 | 2024-03-19 | Power control method, system and equipment based on source network charge storage |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118336829A (en) |
-
2024
- 2024-03-19 CN CN202410311320.2A patent/CN118336829A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ye et al. | Predicting electricity consumption in a building using an optimized back-propagation and Levenberg–Marquardt back-propagation neural network: Case study of a shopping mall in China | |
| Rahman et al. | Predicting electricity consumption for commercial and residential buildings using deep recurrent neural networks | |
| Guan et al. | An artificial-neural-network-based, constrained CA model for simulating urban growth | |
| CN116993532B (en) | Method and system for improving preparation efficiency of battery parts | |
| CN115759413B (en) | Meteorological prediction method and device, storage medium and electronic equipment | |
| CN117114438A (en) | Building area energy system cold and hot load data driving prediction method with flexibility and interpretability | |
| CN114219023A (en) | Data clustering method and device, electronic equipment and readable storage medium | |
| CN115829124A (en) | Charging pile address selection method, device, equipment and storage medium | |
| CN115310727A (en) | Building cooling, heating and power load prediction method and system based on transfer learning | |
| CN112465141A (en) | Model compression method, model compression device, electronic device and medium | |
| CN117155771A (en) | Equipment cluster fault tracing method and device based on industrial Internet of things | |
| CN118336829A (en) | Power control method, system and equipment based on source network charge storage | |
| CN116401602A (en) | Event detection method, device, equipment and computer readable medium | |
| CN114997052A (en) | Fan service life prediction method, device, equipment and medium based on separable convolution | |
| Cho et al. | Application of Parallel ANN-PSO to Hourly Solar PV Estimation | |
| CN117706045B (en) | Combined control method and system for realizing atmospheric ozone monitoring equipment based on Internet of things | |
| CN118336780B (en) | Capacity configuration method, device and equipment of photovoltaic energy storage system and storage medium | |
| CN116957854A (en) | Intelligent configuration method, device, equipment and storage medium for equipment leasing network | |
| CN118821506A (en) | Discrete fracture-cavity unstructured grid optimization method and device, electronic equipment and medium | |
| CN118821505A (en) | Method, device, electronic equipment and medium for determining unstructured grid control node | |
| CN118572699B (en) | Charging pile charging charge prediction method, system and device based on Internet of things | |
| CN118783533A (en) | Distributed power supply access and control method and system in intelligent circuit breaker | |
| CN117435870B (en) | Load data real-time filling method, system, equipment and medium | |
| CN116522105B (en) | Method, device, equipment and medium for integrally constructing data based on cloud computing | |
| Indralaksono et al. | Hierarchical Clustering and Deep Learning for Short-Term Load Forecasting with Influenced Factors |
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 |