CN117689188A - Big data-based user charging strategy optimization system and method - Google Patents
Big data-based user charging strategy optimization system and method Download PDFInfo
- Publication number
- CN117689188A CN117689188A CN202410157084.3A CN202410157084A CN117689188A CN 117689188 A CN117689188 A CN 117689188A CN 202410157084 A CN202410157084 A CN 202410157084A CN 117689188 A CN117689188 A CN 117689188A
- Authority
- CN
- China
- Prior art keywords
- charging
- user
- analyzed
- feature
- energy price
- 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.)
- Granted
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 58
- 238000000034 method Methods 0.000 title claims abstract description 42
- 239000013598 vector Substances 0.000 claims description 179
- 238000000605 extraction Methods 0.000 claims description 53
- 239000011159 matrix material Substances 0.000 claims description 21
- 238000013527 convolutional neural network Methods 0.000 claims description 20
- 238000013508 migration Methods 0.000 claims description 13
- 230000005012 migration Effects 0.000 claims description 13
- 238000009826 distribution Methods 0.000 claims description 12
- 230000004927 fusion Effects 0.000 claims description 5
- 238000012545 processing Methods 0.000 description 14
- 238000010586 diagram Methods 0.000 description 12
- 238000004458 analytical method Methods 0.000 description 11
- 230000008859 change Effects 0.000 description 8
- 238000011161 development Methods 0.000 description 8
- 230000018109 developmental process Effects 0.000 description 8
- 230000006870 function Effects 0.000 description 7
- 230000000875 corresponding effect Effects 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 230000004913 activation Effects 0.000 description 4
- 230000006399 behavior Effects 0.000 description 4
- 238000011176 pooling Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000005611 electricity Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000003062 neural network model Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000012546 transfer Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000001276 controlling effect Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000012098 association analyses Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 108010015046 cell aggregation factors Proteins 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000001627 detrimental effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 239000000295 fuel oil Substances 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000015654 memory Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to the field of user charging strategy optimization, and particularly discloses a user charging strategy optimization system and method based on big data. In this way, a proper charging strategy can be provided according to the requirements of users, and the usability and user experience of the charging system are improved.
Description
Technical Field
The present application relates to the field of policy optimization for user charging, and more particularly, to a system and method for user charging policy optimization based on big data.
Background
With the tension of energy supply and increasing attention to environmental protection, electric automobiles are widely popularized and applied as a clean and efficient transportation means. The method can reduce the dependence on the traditional fuel oil and reduce the influence of exhaust emission on air quality and environment. With the rapid development of new energy electric vehicles, the number of electric vehicles is exponentially increased nowadays, which results in a rapid increase in charging demand. Although the construction of charging stations partially alleviates the problem of charging electric vehicles, the challenges of unbalanced supply and demand of charging piles are also increasingly apparent. In addition, the charging needs of different users are different. Some users may need to be charged quickly to meet the needs of emergency vehicles, while some users may be more concerned with controlling the cost of charging, and may wish to charge during periods of lower power prices.
Therefore, a user charging strategy optimization system and method based on big data are expected, and through real-time acquisition and analysis of user charging data, energy price and charging pile state, the personalized, cost-optimized, efficient and energy-saving optimal charging strategy is improved for the user, and the availability and user experience of the charging system are improved.
Disclosure of Invention
The present application has been made in order to solve the above technical problems. The embodiment of the application provides a user charging strategy optimization system and method based on big data, which comprises the steps of firstly collecting user charging data to be analyzed, energy price to be analyzed and charging pile state to be analyzed, then carrying out context semantic understanding on the user charging data to be analyzed through a context encoder, carrying out feature extraction on the energy price to be analyzed through a convolutional neural network model, carrying out feature analysis on the charging pile state to be analyzed through a time sequence encoder and a multi-scale convolutional structure in sequence, and then generating a user charging strategy by combining the feature information. In this way, a proper charging strategy can be provided according to the requirements of users, and the usability and user experience of the charging system are improved.
According to a first aspect of the present application, there is provided a big data based user charging policy optimization system comprising:
the user data acquisition module is used for acquiring user charging data to be analyzed, energy price to be analyzed and state of the charging pile to be analyzed;
the user characteristic acquisition module is used for acquiring characteristic information of the user charging data to be analyzed, characteristic information of the energy price to be analyzed and characteristic information of the charging pile state to be analyzed so as to obtain a user charging-energy price association characteristic vector and a multi-scale charging pile state characteristic vector;
and the user strategy generation module is used for generating a user charging strategy suitable for the user to be analyzed based on the user charging-energy price correlation feature vector and the multi-scale charging pile state feature vector.
With reference to the first aspect of the present application, in a user charging policy optimization system based on big data in the first aspect of the present application, the user feature obtaining module includes: the user charging feature extraction unit is used for carrying out convolutional encoding on the user charging data to be analyzed to obtain user charging feature vectors; the energy feature extraction unit is used for carrying out convolutional encoding on the energy price to be analyzed to obtain an energy price feature vector; the feature fusion unit is used for obtaining a user charging-energy price association feature vector by combining the user charging feature vector and the energy price feature vector; and the charging pile state feature extraction unit is used for carrying out convolutional encoding on the charging pile state to be analyzed so as to obtain the multi-scale charging pile state feature vector.
According to a second aspect of the present application, there is provided a user charging policy optimization method based on big data, comprising:
collecting charging data of a user to be analyzed, energy price to be analyzed and state of a charging pile to be analyzed;
acquiring characteristic information of the user charging data to be analyzed, characteristic information of the energy price to be analyzed and characteristic information of the charging pile state to be analyzed so as to obtain a user charging-energy price correlation characteristic vector and a multi-scale charging pile state characteristic vector;
and generating a user charging strategy suitable for the user to be analyzed based on the user charging-energy price association feature vector and the multi-scale charging pile state feature vector.
With reference to the second aspect of the present application, in a user charging policy optimization method based on big data in the second aspect of the present application, generating a user charging policy suitable for the user to be analyzed based on the user charging-energy price association feature vector and the multi-scale charging pile state feature vector includes: the distance migration unit is used for carrying out target dimension probability density distribution distance migration on the user charging-energy price association feature vector and the multi-scale charging pile state feature vector so as to obtain a user charging strategy feature matrix; and the strategy generation unit is used for enabling the user charging strategy characteristic matrix to pass through a generator to obtain a user charging strategy suitable for the user to be analyzed.
Compared with the prior art, the user charging strategy optimization system and method based on big data are characterized in that firstly user charging data to be analyzed, energy price to be analyzed and charging pile state to be analyzed are collected, then context semantic understanding is conducted on the user charging data to be analyzed through a context encoder, feature extraction is conducted on the energy price to be analyzed through a convolutional neural network model, feature analysis is conducted on the charging pile state to be analyzed through a time sequence encoder and a multi-scale convolutional structure, and then a user charging strategy is generated by combining the feature information. In this way, a proper charging strategy can be provided according to the requirements of users, and the usability and user experience of the charging system are improved.
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 illustrates a schematic block diagram of a big data based user charging policy optimization system in accordance with an embodiment of the present application.
Fig. 2 illustrates a schematic block diagram of a user feature acquisition module in a big data based user charging policy optimization system in accordance with an embodiment of the present application.
Fig. 3 illustrates a schematic block diagram of a user charging feature extraction unit in a user feature acquisition module in a big data based user charging policy optimization system according to an embodiment of the present application.
Fig. 4 illustrates a schematic block diagram of an energy feature extraction unit in a user feature acquisition module in a big data based user charging policy optimization system according to an embodiment of the present application.
Fig. 5 illustrates a schematic block diagram of a charging pile status feature extraction unit in a user feature acquisition module in a big data based user charging policy optimization system according to an embodiment of the present application.
Fig. 6 illustrates a schematic block diagram of a user policy generation module in a big data based user charging policy optimization system in accordance with an embodiment of the present application.
Fig. 7 illustrates a flowchart of a big data based user charging policy optimization method 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.
Exemplary System
FIG. 1 illustrates a schematic block diagram of a big data based user charging policy optimization system in accordance with an embodiment of the present application. As shown in fig. 1, a big data based user charging policy optimization system 100 according to an embodiment of the present application includes: the user data acquisition module 110 is used for acquiring user charging data to be analyzed, energy price to be analyzed and state of charging pile to be analyzed; the user characteristic obtaining module 120 is configured to obtain characteristic information of the user charging data to be analyzed, characteristic information of the energy price to be analyzed, and characteristic information of the charging pile state to be analyzed, so as to obtain a user charging-energy price association characteristic vector and a multi-scale charging pile state characteristic vector; the user policy generation module 130 is configured to generate a user charging policy suitable for the user to be analyzed based on the user charging-energy price association feature vector and the multi-scale charging pile state feature vector.
The development of the electric automobile brings convenience such as environmental friendliness, energy diversification and cost saving, and the construction of the charging pile provides the electric automobile with the advantages of convenient charging, charging network development, technical innovation and the like, and promotes the popularization and development of the electric automobile.
However, as described in the background section above, the charging needs of different users also vary. Some users may need to be charged quickly to meet the needs of emergency vehicles, while some users may be more concerned with controlling the cost of charging, and may wish to charge during periods of lower power prices. Therefore, a user charging strategy optimization system and method based on big data are expected, and through real-time acquisition and analysis of user charging data, energy price and charging pile state, the personalized, cost-optimized, efficient and energy-saving optimal charging strategy is improved for the user, and the availability and user experience of the charging system are improved.
It should be appreciated that the development of large data is due to a combination of factors such as the growth of data volume, data diversity, advances in data processing technology, data mining, and the application of machine learning. The development of big data provides more data resources and analysis tools for us, and brings new opportunities and challenges for innovation and development of various fields. In the embodiment of the application, the development of big data provides the advantages of big data volume, data diversity, instantaneity and instantaneity, prediction and optimization capability, data driving decision-making and the like for the optimization of the charging strategy, and can help to improve the performance and user experience of the charging system. Therefore, the input data in the embodiments of the present application are analyzed by big data technique.
In this embodiment of the present application, the user data collection module 110 is configured to collect user charging data to be analyzed, energy price to be analyzed, and charging pile status to be analyzed. It should be appreciated that the user charging data, energy price and charging pile status are key information for developing the charging strategy. By collecting and analyzing the charging data of the user, the charging habit, the driving mode, the charging requirement and the like of the user can be known, so that a personalized charging strategy is provided for the user. Meanwhile, the energy price data are collected and analyzed, so that the change trend of the energy price can be predicted, and the user is helped to select the optimal charging period. The charging pile state data can reflect the service condition, the power supply capacity and the like of the charging pile, and real-time resource information is provided for the formulation of the charging strategy.
Specifically, when collecting the user data that charges that wait to analyze, can install the sensor or use intelligent charging stake on charging stake, can gather user's data that charges in real time. Such data may include charge start and end times, charge power, charge level, charge duration, etc. The sensor can record state information of the charging pile, such as the occupied state, the power supply capacity and the like of the charging pile. The intelligent charging pile can send data to a data center for analysis through network connection. In addition, when the energy price is to be analyzed, the energy price data can be obtained in real time by accessing a website of the energy market or using a related API (application program interface). Such data may include real-time electricity prices of the electricity market, electricity price variations of different periods, and the like. Further, when the state of the charging pile to be analyzed is obtained, a sensor can be installed on the charging pile and used for collecting state information of the charging pile. The sensor can monitor parameters such as current, voltage, temperature and the like of the charging pile, and transmits data to the data acquisition system for analysis. Through the installation and data acquisition of sensor, can acquire the state data of filling electric pile in real time.
In this embodiment of the present application, the user feature obtaining module 120 is configured to obtain feature information of the user charging data to be analyzed, feature information of the energy price to be analyzed, and feature information of the charging pile state to be analyzed, so as to obtain a user charging-energy price associated feature vector and a multi-scale charging pile state feature vector. It should be appreciated that after the input data is collected, the input data is further subjected to feature extraction and analysis tasks.
In particular, fig. 2 illustrates a schematic block diagram of a user feature acquisition module in a big data based user charging policy optimization system according to an embodiment of the present application. As shown in fig. 2, the user feature obtaining module 120 includes: a user charging feature extraction unit 121, configured to convolutionally encode the user charging data to be analyzed to obtain the user charging feature vector; an energy feature extraction unit 122, configured to convolutionally encode the energy price to be analyzed to obtain the energy price feature vector; a feature fusion unit 123, configured to combine the user charging feature vector and the energy price feature vector to obtain a user charging-energy price association feature vector; and the charging pile state feature extraction unit 124 is configured to convolutionally encode the charging pile state to be analyzed to obtain the multi-scale charging pile state feature vector.
And firstly, extracting and analyzing the characteristics of the charging data of the user to be analyzed. Specifically, fig. 3 illustrates a schematic block diagram of a user charging feature extraction unit in a user feature acquisition module in a big data based user charging policy optimization system according to an embodiment of the present application. As shown in fig. 3, the user charging feature extraction unit 121 includes: the user charging data embedding encoding subunit 121-1 is configured to perform embedding encoding on the user charging data to be analyzed to obtain a user input vector; and the user charging feature acquisition subunit 121-2 is configured to obtain the user charging feature vector by using the user charging input vector through a user charging feature extraction module based on a one-dimensional convolutional neural network model.
It should be appreciated that in this system, the user charging data to be analyzed includes a user's charging peg usage record, a charging start time, a charging end time, a charging duration, and power, etc. These text data are difficult to directly understand and extract features by the machine. Thus, these data are first embedded encoded to convert them into a machine recognizable vector form. Further, it is considered that the user charging data includes not only the charging stake use record but also various types of data such as a charging start time, a charging end time, and the like. The Clip model is able to process multimodal data and map them into the same vector space. Therefore, in the embodiment of the application, the Clip model is used for embedded encoding of the user charging data to be analyzed. Thus, the charging characteristics of the user can be more comprehensively captured, and the accuracy and reliability of the charging strategy analysis of the user are improved.
In a specific embodiment of the present application, the user charging data is embedded in the encoding subunit 121-1 for: word segmentation processing is carried out on the user charging data to be analyzed to obtain word sequences; mapping each word in the word sequence into a word embedding vector by using an embedding layer of a sequence encoder of the Clip model to obtain a sequence of word embedding vectors; performing global-based context semantic coding on the sequence of the embedded vectors by using a converter-based Bert model of a sequence encoder of the Clip model to obtain a plurality of feature vectors; and concatenating the plurality of feature vectors to obtain the user input vector.
Further, considering that convolutional neural networks have local perceptibility, local patterns and features in the input vector can be identified through convolutional operations. In the user charging data, there may be some local characteristic patterns, such as current waveforms, power variations during charging, etc. These local features can be captured by the convolutional neural network model through a convolutional operation, thereby extracting useful user charging features. Thus, the user input vector is further subjected to high-dimensional local feature extraction through a convolutional neural network model.
In a specific embodiment of the present application, the user charging feature acquisition subunit 121-2 is configured to: each layer of the one-dimensional convolutional neural network model is used for respectively carrying out forward transfer on input data in the layers: performing convolution processing based on a one-dimensional convolution kernel on the input data to obtain a convolution feature map; pooling processing is carried out on the convolution feature images based on feature matrixes to obtain pooled feature images; non-linear activation is carried out on the pooled feature map so as to obtain an activated feature map; the output of the last layer of the one-dimensional convolutional neural network model is a user charging feature vector.
And then extracting and analyzing the characteristics of the energy price to be analyzed. Specifically, fig. 4 illustrates a schematic block diagram of an energy feature extraction unit in a user feature acquisition module in a big data based user charging policy optimization system according to an embodiment of the present application. As shown in fig. 4, the energy feature extraction unit 122 includes: an energy data arrangement subunit 122-1, configured to arrange the energy price to be analyzed into an energy price input vector; the energy feature obtaining subunit 122-2 is configured to pass the energy price input vector through an energy price feature extraction module based on a convolutional neural network model to obtain the energy price feature vector.
It should be appreciated that changes in energy prices typically have certain contextual information, such as trends in price, periodic changes, and the like. The energy prices are arranged as the input vectors, so that the context information can be better captured and fused with other input factors, and a more accurate and personalized charging strategy is provided. Therefore, before the feature extraction is performed on the energy price to be analyzed, the energy price to be analyzed is firstly arranged to obtain the energy price input vector.
Further, convolutional neural networks (Convolutional Neural Network, CNN) are considered to be a neural network model that is widely used in the fields of computer vision and image processing. The method can effectively capture the spatial local features in the input data, and perform feature extraction and dimension reduction through layer-by-layer convolution and pooling operations. And the energy price input vector may contain a large amount of time series data, such as energy prices for different time periods. By using the energy price characteristic extraction module based on the convolutional neural network model, important characteristic information can be extracted from the energy price input vector to obtain an energy price characteristic vector. That is, the convolutional neural network may capture local features in the energy price input vector through a convolutional operation. This is important for fluctuations and patterns of energy prices, which can help the system to better understand the trends and periodicity of energy prices. Therefore, in order to better acquire important feature information of the energy price, the energy price input vector is subjected to deep convolution coding by an energy price feature extraction module based on a convolution neural network model so as to obtain an energy price feature vector with more feature representation.
In one embodiment of the present application, the energy feature acquisition subunit 122-2 is configured to: each layer using the convolutional neural network model performs the following steps on input data in forward transfer of the layer: performing convolution processing based on a one-dimensional convolution kernel on the input data to obtain a convolution feature map; pooling processing is carried out on the convolution feature images based on feature matrixes to obtain pooled feature images; non-linear activation is carried out on the pooled feature map so as to obtain an activated feature map; the input of the first layer of the convolutional neural network model is the energy price input vector, and the output of the last layer of the convolutional neural network model is the energy price characteristic vector.
Further, it is considered that there is a certain association between the charging behavior of the user and the energy price, because the change of the energy price affects the charging behavior of the user and the policy selection. For example, when the energy price is low, the user may prefer to charge during the valley period to obtain a lower charge cost. When the energy price is high, the user may adjust the charging behavior to select to charge in a period of low energy price, so as to reduce the charging cost. Thus, to obtain these correlation features, the user charging feature vector and the energy price feature vector are combined to obtain a user charging-energy price correlation feature vector. Therefore, by analyzing the relation between the user charging feature vector and the energy price feature vector, the system can provide personalized charging suggestions and optimization strategies according to the current energy price condition and the charging requirement of the user so as to meet the requirement of the user to the greatest extent and reduce the charging cost.
In a specific embodiment of the present application, the feature fusion unit 123 is configured to: fusing the user charging feature vector and the energy price feature vector by using a cascading formula to obtain a user charging-energy price association feature vector; wherein, the cascade formula is:
wherein,representing the user charge-energy price associated feature vector,representing the energy price characteristic vector,representing the user charge-energy price associated feature vector,representing a cascading function.
And then extracting and analyzing the characteristics of the state of the charging pile to be analyzed. Specifically, fig. 5 illustrates a schematic block diagram of a charging pile state feature extraction unit in a user feature acquisition module in a big data based user charging policy optimization system according to an embodiment of the present application. As shown in fig. 5, the charging pile state feature extraction unit 124 includes: the charging pile state embedding and encoding subunit 124-1 is configured to perform embedding and encoding on the charging pile states of each charging pile in the charging pile states to be analyzed so as to obtain a plurality of charging pile state input vectors; a charging pile time sequence feature extraction subunit 124-2, configured to pass the plurality of charging pile state input vectors through a charging pile state feature extraction module based on a time sequence encoder to obtain a plurality of charging pile state feature vectors; and the charging pile multi-scale state feature extraction subunit 124-3 is configured to pass the plurality of charging pile state feature vectors through a charging pile state feature extraction module with a multi-scale convolution structure to obtain the multi-scale charging pile state feature vectors. Wherein the timing encoder includes a full-connection layer and a one-dimensional convolutional layer.
It should be understood that in this system, the state of the charging pile to be analyzed includes the working state of the charging pile, such as charging, idle, failure, etc.; the connection condition of the charging pile and the power grid, such as connection, disconnection and the like, and the charging rate or power output condition of the charging pile. These text data are difficult to directly understand and extract features by the machine. Thus, these data are first embedded encoded to convert them into a machine recognizable vector form. Further, it is considered that the charging pile status includes not only the charging pile usage record but also various types of data such as a charging start time, a charging end time, and the like. The Clip model is able to process multimodal data and map them into the same vector space. Therefore, in the embodiment of the present application, the Clip model is used to perform embedded encoding on the charging pile state. Thus, the charging characteristics of the user can be more comprehensively captured, and the accuracy and reliability of the state analysis of the charging pile are improved.
In a specific embodiment of the present application, the charging pile state embedding encoding subunit 124-1 is configured to: word segmentation processing is respectively carried out on the charging pile states of all charging piles in the charging pile states to be analyzed so as to obtain word sequences; mapping each word in the word sequence into a word embedding vector by using an embedding layer of a sequence encoder of the Clip model to obtain a sequence of word embedding vectors; performing global-based context semantic coding on the sequence of the embedded vectors by using a converter-based Bert model of a sequence encoder of the Clip model to obtain a plurality of feature vectors; and cascading the plurality of feature vectors to obtain the plurality of charging pile state input vectors.
It should be appreciated, in turn, that the state of charge stake contains information in a number of ways, such as the operational state of the charge stake, the connection state, the rate of charge, etc., which information is characteristic of a time sequence change. For example, at different points in time, the operational state of the charging stake may be on or off, the connection state of the charging stake may be on or off, etc. Considering that a time series encoder is a neural network model capable of processing time series data, common time series encoders include a cyclic neural network (Recurrent Neural Network, RNN) and a Transformer (transducer). The models can capture time sequence dependency and long-term memory in time sequence data, and are very effective for processing time sequence data of the state of the charging pile. Thus, the characteristics of the state of the charging pile in the time dimension are extracted and analyzed by using the time-series encoder-based charging pile state characteristic extraction module. In this way, an understanding of the change and evolution of the state of the charging pile is facilitated.
In a specific embodiment of the present application, the charging pile timing feature extraction subunit 124-2 is configured to: performing full-connection coding on the input vector by using a full-connection layer of the time sequence coder to extract high-dimensional implicit features of feature values of all positions in the input vector; and performing one-dimensional convolutional encoding on the input vector by using a one-dimensional convolutional layer of a time sequence encoder to extract associated high-dimensional implicit association features among feature values of all positions in the input vector so as to obtain the state feature vectors of the charging piles.
Further, it should be understood that there is multi-scale feature information between different charging piles, i.e. on different time scales, the states of different charging piles may exhibit different variation patterns and features. And the multi-scale characteristics of different charging piles are subjected to association analysis, so that the method is beneficial to selecting more proper charging piles for users. Based on the multi-scale feature extraction is performed on the plurality of charging pile state feature vectors by a charging pile state feature extraction module with a multi-scale convolution structure. In this way, by the feature extraction module with the multi-scale convolution structure, the state features of the charging pile can be extracted and modeled on different time scales. Thus, the change trend and mode of the state of the charging pile on different time scales can be captured, and thus more comprehensive and rich characteristic representation is provided.
In a specific embodiment of the present application, the charging pile multi-scale state feature extraction subunit 124-3 is configured to: each layer of the charging pile state characteristic extraction module with the multi-scale convolution structure is used for respectively carrying out input data in the forward transfer process of the layer: performing convolution processing based on a first convolution kernel on the input data to obtain a first scale convolution feature map; performing convolution processing based on a second convolution kernel on the input data to obtain a second scale convolution feature map; performing convolution processing based on a third convolution kernel on the input data to obtain a third scale convolution feature map; cascading the first scale convolution feature map, the second scale convolution feature map and the third scale convolution feature map to obtain a multi-scale convolution feature map; pooling the multi-scale convolution feature map to obtain a pooled feature map; performing nonlinear activation on the pooled feature map to obtain a nonlinear activation feature map; the output of the last layer of the charging pile state feature extraction module with the multi-scale convolution structure is the multi-scale charging pile state feature vector.
In this embodiment of the present application, the user policy generation module 130 is configured to generate a user charging policy suitable for the user to be analyzed based on the user charging-energy price association feature vector and the multi-scale charging pile state feature vector.
In particular, fig. 6 illustrates a schematic block diagram of a user policy generation module in a big data based user charging policy optimization system in accordance with an embodiment of the present application. As shown in fig. 6, the user policy generation module 130 includes: the distance migration unit 131 is configured to perform target dimension probability density distribution distance migration on the user charging-energy price association feature vector and the multi-scale charging pile state feature vector to obtain a user charging policy feature matrix; and the policy generation unit 132 is configured to pass the user charging policy feature matrix through a generator to obtain a user charging policy suitable for the user to be analyzed.
Particularly, in the technical scheme of the application, when the user charging-energy price association feature vector and the multi-scale charging pile state feature vector are fused, baseline drift may be caused due to the influence of noise signals, so that unnecessary negative correlation features are introduced into the user charging strategy feature matrix. Baseline wander is a change in the overall trend of the data, possibly due to the presence of noise signals. In the technical scheme, the user charging-energy price correlation feature vector and the multi-scale charging pile state feature vector are used for constructing a user charging strategy feature matrix. However, the noise signal may disturb the correlation that is originally present, resulting in a change in the overall trend of the data, thereby introducing unnecessary negative correlation features. Noise signals may come from a variety of factors, such as errors in the user's charge data acquisition process, fluctuations in energy prices, measurement errors in the charge pile state acquisition, and the like. In the technical scheme, the factors can cause the real association relation between the user charging-energy price association feature vector and the multi-scale charging pile state feature vector to change. For example, there may be a positive correlation between the user charge-energy price correlation feature vector and the multi-scale charging pile state feature vector at a certain point in time, but due to the presence of noise signals, there may be cases where the user charge-energy price correlation feature vector increases and the multi-scale charging pile state feature vector decreases, thereby generating incorrect negative correlation features in the user charge policy feature matrix. Baseline wander is a manifestation of noise signals that represents changes in the overall trend of the data. When the baseline drifts, the correlation that is originally present may be altered, thereby introducing unnecessary negative correlation features. For example, the user charging-energy price correlation feature vector and the multi-scale charging pile state feature vector at a certain time point have a positive correlation relationship in a certain time period, but the overall trend of data changes due to the influence of baseline drift, and the multi-scale charging pile state feature vector is reduced while the user charging-energy price correlation feature vector is increased, so that incorrect negative correlation features appear in the user charging strategy feature matrix, and the accuracy of generating the user charging strategy to be analyzed is negatively influenced.
In a specific embodiment of the present application, the distance migration unit 131 is configured to: performing target dimension probability density distribution distance migration on the user charging-energy price association feature vector and the multi-scale charging pile state feature vector by using the following formula to obtain a user charging strategy feature matrix; wherein, the formula is:
wherein,representing the user charge-energy price associated feature vector,representing the multi-scale charging pile state characteristic vector,the transpose of the vector is represented,the vector multiplication is represented by a vector,a linear rectification function is represented and is used,a Frobenius norm representing the user charge-energy price correlation feature vector,a Frobenius norm representing the multi-scale charging pile state feature vector,and representing the user charging strategy feature matrix.
That is, when the user charge-energy price related feature vector and the multi-scale charge pile state feature vector are fused, there may be a baseline drift due to the influence of noise signals, that is, the feature distribution and feature shape of the two feature vectors in the high-dimensional feature space may be shifted or changed due to the interference of the noise signals, which results in the introduction of unnecessary negative correlation features in the user charge policy feature matrix as the user charge-energy price related feature vector and the multi-scale charge pile state feature vector, that is, the feature distribution and feature shape of the two feature vectors in the high-dimensional feature space may be negatively correlated due to the interference of the noise signals, which are detrimental to the generation task, because they may affect the judgment and decision of the generator, thereby reducing the accuracy and efficiency of the generation.
In order to solve the problem, the application provides a method for performing target dimension probability density distribution distance migration on the user charging-energy price correlation feature vector and the multi-scale charging pile state feature vector to obtain a user charging strategy feature matrix, namely, each feature vector can obtain a target dimension probability density distribution distance migration according to the distance between the feature vector and the probability density distribution generating the target dimension by using the target dimension-based method. Specifically, the method is based on a ReLU function, so that each eigenvector can obtain a ReLU function value according to the positive and negative of projection. And each feature vector can be obtained into a nonlinear re-weighting mechanism according to the distribution of the feature vector and projection re-weighting by a method based on the nonlinear re-weighting mechanism. That is, the user charging-energy price association feature vector and the multi-scale charging pile state feature vector can both obtain a space aggregation factor according to the space aggregation of the user charging-energy price association feature vector and different association features. In this way, feature inconsistencies caused by different dimensions and different directions between feature vectors can be effectively eliminated or reduced by implementing the target dimension probability density distribution distance migration of the feature vectors.
Further, it should be understood that the user charging policy feature matrix includes the entire deep feature information related to user charging, so as to provide more comprehensive reference information for generating the user charging policy. The generator can learn and analyze according to the information such as the charging strategy feature matrix of the user, historical data and the like, so that a charging strategy based on data is generated. Thus, the charging demand and behavior of the user can be predicted more accurately by using a data-driven method, and a corresponding charging strategy is generated. In this way, a more appropriate and intended charging strategy can be provided for the user, and the usability and user experience of the charging system are further improved.
In a specific embodiment of the present application, the policy generating unit 132 specifically performs the policy generating process as follows: 1. data preprocessing: firstly, carrying out data preprocessing on a charging strategy feature matrix of a user to be analyzed, wherein the data preprocessing comprises the steps of data cleaning, feature selection, standardization and the like. These steps can ensure the quality and consistency of the input data, providing a high quality input to the generator. 2. The generator selects: an appropriate generator method is selected based on system requirements and algorithm selection. Common generators include optimization algorithm generators, rule-based generators, and machine learning generators. The selection of the appropriate generator method may be determined based on the particular problem and data characteristics. 3. Generator training or configuration: for machine learning generators, a training phase is required to be performed, and a generator model is trained using a charging strategy feature matrix of the user to be analyzed as input. For an optimization algorithm generator or rule-based generator, it may be necessary to configure the corresponding parameters and rules so that the generator can generate the corresponding charging strategy from the input. 4. And (3) strategy generation: and according to the generator method and the trained model, taking a charging strategy feature matrix of the user to be analyzed as input to generate a corresponding charging strategy. The generated strategy can comprise specific decision contents such as charging time period, power limit, charging equipment selection and the like so as to meet the personalized requirements of users and the optimization targets of the system. 5. Policy evaluation and adjustment: the generated charging strategy can be evaluated and verified to ensure that the charging strategy meets the requirements of users and the constraints of the system. If further optimization or adjustment of the strategy is required, iterations and improvements can be made, re-running the generator or adjusting the parameters of the generator, to get better strategy results.
It should be noted that the specific policy generation process may vary depending on the system requirements and the generator method. The above is a general flow example, and in practical application, adjustment and optimization are required according to specific situations.
It is worth mentioning that instead of using a generator, the generation may be based on expert knowledge and rules. This approach relies on the experience and knowledge of domain experts and generates a charging strategy by defining a series of rules and logic.
The following are general steps of a policy generation scheme based on expert knowledge and rules: 1. domain expert knowledge acquisition: communicate and cooperate with experts in the charging field to acquire their experience and knowledge. These professionals may be power engineers, energy management professionals, or other professionals in the relevant arts. 2. Rule definition: based on expert knowledge, a series of rules and logic are defined for generating the charging strategy. These rules may include determining a charging period based on user demand and grid conditions, setting power limits, considering availability of charging devices, and so forth. 3. And (3) strategy generation: and generating a corresponding charging strategy according to the characteristics and the requirements of the user to be analyzed and the defined rules. And gradually applying the rules according to the priority and the conditions of the rules to generate various aspects of the charging strategy. 4. Policy evaluation and adjustment: the generated charging strategy can be evaluated and verified to ensure that the charging strategy meets the requirements of users and the constraints of the system. If further optimization or adjustment strategies are required, appropriate adjustments and improvements can be made with the expert's discussion and feedback.
The generation based on expert knowledge and rules may be strategically generated without extensive data or training samples. It can provide interpretability and controllability depending on the experience and domain knowledge of the expert. However, this approach may not fully exploit the discovery of large data and complex patterns, and thus may not be as flexible and accurate as generator-based methods in some cases. The selection of the appropriate policy generation should be determined based on the specific requirements and available resources.
In summary, the big data based user charging policy optimization system according to the embodiments of the present application is clarified, which collects user charging data to be analyzed, energy price to be analyzed and charging pile state to be analyzed first, then performs context semantic understanding on the user charging data to be analyzed through a context encoder, performs feature extraction on the energy price to be analyzed through a convolutional neural network model, and performs feature analysis on the charging pile state to be analyzed sequentially through a time sequence encoder and a multi-scale convolutional structure, and then generates a user charging policy in combination with these feature information. In this way, a proper charging strategy can be provided according to the requirements of users, and the usability and user experience of the charging system are improved.
As described above, the big data based user charging policy optimization system 100 according to the embodiment of the present application may be implemented in various wireless terminals, such as a server or the like for big data based user charging policy optimization. In one example, big data based user charging policy optimization system 100 according to embodiments of the present application may be integrated into a wireless terminal as one software module and/or hardware module. For example, the big data based user charging policy optimization system 100 may be a software module in the operating system of the wireless terminal or may be an application developed for the wireless terminal; of course, the big data based user charging policy optimization system 100 could equally be one of many hardware modules of the wireless terminal.
Alternatively, in another example, the big data based user charging policy optimization system 100 and the wireless terminal may be separate devices, and the big data based user charging policy optimization system 100 may be connected to the wireless terminal through a wired and/or wireless network and transmit the interaction information in a agreed data format.
Exemplary method
Fig. 7 illustrates a flowchart of a big data based user charging policy optimization method according to an embodiment of the present application. As shown in fig. 7, the big data based user charging policy optimization method according to the embodiment of the present application includes: s1, collecting charging data of a user to be analyzed, energy price to be analyzed and state of a charging pile to be analyzed; s2, obtaining characteristic information of the user charging data to be analyzed, characteristic information of the energy price to be analyzed and characteristic information of the charging pile state to be analyzed so as to obtain a user charging-energy price correlation characteristic vector and a multi-scale charging pile state characteristic vector; and S3, generating a user charging strategy suitable for the user to be analyzed based on the user charging-energy price association feature vector and the multi-scale charging pile state feature vector.
Here, it will be understood by those skilled in the art that the specific functions and operations of the respective steps in the above-described big data based user charging policy optimization method have been described in detail in the above description of the big data based user charging policy optimization system with reference to fig. 1, and thus, repetitive descriptions thereof will be omitted.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit 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 functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A big data based user charging policy optimization system, comprising:
the user data acquisition module is used for acquiring user charging data to be analyzed, energy price to be analyzed and state of the charging pile to be analyzed;
the user characteristic acquisition module is used for acquiring characteristic information of the user charging data to be analyzed, characteristic information of the energy price to be analyzed and characteristic information of the charging pile state to be analyzed so as to obtain a user charging-energy price association characteristic vector and a multi-scale charging pile state characteristic vector;
and the user strategy generation module is used for generating a user charging strategy suitable for the user to be analyzed based on the user charging-energy price correlation feature vector and the multi-scale charging pile state feature vector.
2. The big data based user charging policy optimization system of claim 1, wherein the user feature acquisition module comprises:
the user charging feature extraction unit is used for carrying out convolutional encoding on the user charging data to be analyzed to obtain user charging feature vectors;
the energy feature extraction unit is used for carrying out convolutional encoding on the energy price to be analyzed to obtain an energy price feature vector;
The feature fusion unit is used for obtaining a user charging-energy price association feature vector by combining the user charging feature vector and the energy price feature vector;
and the charging pile state feature extraction unit is used for carrying out convolutional encoding on the charging pile state to be analyzed so as to obtain the multi-scale charging pile state feature vector.
3. The big data based user charging policy optimization system of claim 2, wherein the user charging feature extraction unit comprises:
the user charging data embedding coding subunit is used for carrying out embedding coding on the user charging data to be analyzed to obtain a user input vector;
and the user charging characteristic acquisition subunit is used for obtaining the user charging characteristic vector by passing the user charging input vector through a user charging characteristic extraction module based on a one-dimensional convolutional neural network model.
4. The big data based user charging policy optimization system of claim 3, wherein the energy feature extraction unit comprises:
the energy data arrangement subunit is used for arranging the energy price to be analyzed into an energy price input vector;
and the energy price characteristic acquisition subunit is used for obtaining the energy price characteristic vector by passing the energy price input vector through an energy price characteristic extraction module based on a convolutional neural network model.
5. The big data based user charging policy optimization system of claim 4, wherein the charging pile state feature extraction unit comprises:
the charging pile state embedding coding subunit is used for carrying out embedding coding on the charging pile states of all charging piles in the charging pile states to be analyzed so as to obtain a plurality of charging pile state input vectors;
the charging pile time sequence feature extraction subunit is used for enabling the plurality of charging pile state input vectors to pass through a charging pile state feature extraction module based on a time sequence encoder so as to obtain a plurality of charging pile state feature vectors;
and the charging pile multi-scale state feature extraction subunit is used for enabling the plurality of charging pile state feature vectors to pass through a charging pile state feature extraction module with a multi-scale convolution structure so as to obtain the multi-scale charging pile state feature vectors.
6. The big data based user charging policy optimization system of claim 5, wherein the timing encoder comprises a full connectivity layer and a one-dimensional convolution layer.
7. The big data based user charging policy optimization system of claim 6, wherein the user policy generation module comprises:
The distance migration unit is used for carrying out target dimension probability density distribution distance migration on the user charging-energy price association feature vector and the multi-scale charging pile state feature vector so as to obtain a user charging strategy feature matrix;
and the strategy generation unit is used for enabling the user charging strategy characteristic matrix to pass through a generator to obtain a user charging strategy suitable for the user to be analyzed.
8. The big data based user charging policy optimization system of claim 7, wherein the distance migration unit is configured to: performing target dimension probability density distribution distance migration on the user charging-energy price association feature vector and the multi-scale charging pile state feature vector by using the following formula to obtain a user charging strategy feature matrix; wherein, the formula is:
wherein,representing the user charging-energy price associated feature vector +.>Representing the state feature vector of the multi-scale charging pile, < >>Representing the transpose of the vector>Representing vector multiplication>Representing a linear rectification function>Frobenius norm representing the user charge-energy price related feature vector, ++>Frobenius norm representing the state feature vector of the multi-scale charging pile,/- >And representing the user charging strategy feature matrix.
9. The user charging strategy optimization method based on big data is characterized by comprising the following steps of:
collecting charging data of a user to be analyzed, energy price to be analyzed and state of a charging pile to be analyzed;
acquiring characteristic information of the user charging data to be analyzed, characteristic information of the energy price to be analyzed and characteristic information of the charging pile state to be analyzed so as to obtain a user charging-energy price correlation characteristic vector and a multi-scale charging pile state characteristic vector;
and generating a user charging strategy suitable for the user to be analyzed based on the user charging-energy price association feature vector and the multi-scale charging pile state feature vector.
10. The method for optimizing a user charging policy based on big data according to claim 9, wherein obtaining the feature information of the user charging data to be analyzed, the feature information of the energy price to be analyzed, and the feature information of the charging pile state to be analyzed to obtain a user charging-energy price correlation feature vector and a multi-scale charging pile state feature vector, comprises:
the user charging feature extraction unit is used for carrying out convolutional encoding on the user charging data to be analyzed to obtain the user charging feature vector;
The energy feature extraction unit is used for carrying out convolutional encoding on the energy price to be analyzed to obtain the energy price feature vector;
the feature fusion unit is used for obtaining a user charging-energy price association feature vector by combining the user charging feature vector and the energy price feature vector;
and the charging pile state feature extraction unit is used for carrying out convolutional encoding on the charging pile state to be analyzed so as to obtain the multi-scale charging pile state feature vector.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410157084.3A CN117689188B (en) | 2024-02-04 | 2024-02-04 | Big data-based user charging strategy optimization system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410157084.3A CN117689188B (en) | 2024-02-04 | 2024-02-04 | Big data-based user charging strategy optimization system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117689188A true CN117689188A (en) | 2024-03-12 |
| CN117689188B CN117689188B (en) | 2024-04-26 |
Family
ID=90137580
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410157084.3A Active CN117689188B (en) | 2024-02-04 | 2024-02-04 | Big data-based user charging strategy optimization system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117689188B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2019033629A (en) * | 2017-08-09 | 2019-02-28 | 三菱重工業株式会社 | Power control method, control device, charging system and program |
| CN109501630A (en) * | 2018-12-04 | 2019-03-22 | 国网电动汽车服务有限公司 | A kind of electric car charging scheme real-time recommendation method and system |
| CN110888908A (en) * | 2019-11-01 | 2020-03-17 | 广州大学 | Charging station/pile recommendation system and method capable of achieving deep learning continuously |
| CN111397620A (en) * | 2020-05-29 | 2020-07-10 | 广东工业大学 | Electric vehicle charging navigation method and system in fast charging/slow charging mode |
| CN111967696A (en) * | 2020-10-23 | 2020-11-20 | 北京国新智电新能源科技有限责任公司 | Neural network-based electric vehicle charging demand prediction method, system and device |
| CN113022334A (en) * | 2019-12-05 | 2021-06-25 | 广西电网有限责任公司 | Remote intelligent charging method and device for electric automobile and storage medium |
| CN114435165A (en) * | 2022-01-25 | 2022-05-06 | 广东电网有限责任公司 | Charging method and device of charging pile, electronic equipment and storage medium |
| CN114998160A (en) * | 2022-08-04 | 2022-09-02 | 江苏游隼微电子有限公司 | Parallel multi-scale feature fusion convolutional neural network denoising method |
| CN114997935A (en) * | 2022-07-19 | 2022-09-02 | 东南大学溧阳研究院 | Electric vehicle charging and discharging strategy optimization method based on interior point strategy optimization |
| WO2023093058A1 (en) * | 2021-11-29 | 2023-06-01 | 华为数字能源技术有限公司 | Charging method and charging apparatus of charging station, and electronic device |
| CN117207803A (en) * | 2023-08-29 | 2023-12-12 | 国网黑龙江省电力有限公司 | Electric automobile intelligent charging strategy selection method based on economic dispatch |
| CN117350424A (en) * | 2023-09-22 | 2024-01-05 | 国网黑龙江省电力有限公司 | Economic dispatching and electric vehicle charging strategy combined optimization method in energy internet |
-
2024
- 2024-02-04 CN CN202410157084.3A patent/CN117689188B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2019033629A (en) * | 2017-08-09 | 2019-02-28 | 三菱重工業株式会社 | Power control method, control device, charging system and program |
| CN109501630A (en) * | 2018-12-04 | 2019-03-22 | 国网电动汽车服务有限公司 | A kind of electric car charging scheme real-time recommendation method and system |
| CN110888908A (en) * | 2019-11-01 | 2020-03-17 | 广州大学 | Charging station/pile recommendation system and method capable of achieving deep learning continuously |
| CN113022334A (en) * | 2019-12-05 | 2021-06-25 | 广西电网有限责任公司 | Remote intelligent charging method and device for electric automobile and storage medium |
| CN111397620A (en) * | 2020-05-29 | 2020-07-10 | 广东工业大学 | Electric vehicle charging navigation method and system in fast charging/slow charging mode |
| CN111967696A (en) * | 2020-10-23 | 2020-11-20 | 北京国新智电新能源科技有限责任公司 | Neural network-based electric vehicle charging demand prediction method, system and device |
| WO2023093058A1 (en) * | 2021-11-29 | 2023-06-01 | 华为数字能源技术有限公司 | Charging method and charging apparatus of charging station, and electronic device |
| CN114435165A (en) * | 2022-01-25 | 2022-05-06 | 广东电网有限责任公司 | Charging method and device of charging pile, electronic equipment and storage medium |
| CN114997935A (en) * | 2022-07-19 | 2022-09-02 | 东南大学溧阳研究院 | Electric vehicle charging and discharging strategy optimization method based on interior point strategy optimization |
| CN114998160A (en) * | 2022-08-04 | 2022-09-02 | 江苏游隼微电子有限公司 | Parallel multi-scale feature fusion convolutional neural network denoising method |
| CN117207803A (en) * | 2023-08-29 | 2023-12-12 | 国网黑龙江省电力有限公司 | Electric automobile intelligent charging strategy selection method based on economic dispatch |
| CN117350424A (en) * | 2023-09-22 | 2024-01-05 | 国网黑龙江省电力有限公司 | Economic dispatching and electric vehicle charging strategy combined optimization method in energy internet |
Non-Patent Citations (4)
| Title |
|---|
| KOOHFAR, S 等: "Performance Comparison of Deep Learning Approaches in Predicting EV Charging Demand", 《SUSTAINABILITY》, vol. 15, no. 5, 31 March 2023 (2023-03-31), pages 2 - 3 * |
| 吴敏: "深度学习框架下充电桩的个性化推荐算法研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》, 15 August 2020 (2020-08-15), pages 3 - 4 * |
| 王鑫 等: "考虑用户因素的电动汽车有序充放电控制策略", 《电力系统保护与控制》, no. 04, 16 February 2018 (2018-02-16), pages 129 - 137 * |
| 贾泽瑞: "考虑用户充电需求及充电站未来客流的充电引导策略研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》, 15 July 2023 (2023-07-15), pages 3 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117689188B (en) | 2024-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111125519B (en) | User behavior prediction method, device, electronic equipment and storage medium | |
| Qin et al. | Next point-of-interest recommendation with auto-correlation enhanced multi-modal transformer network | |
| Huang et al. | Reliable machine prognostic health management in the presence of missing data | |
| CN115206092A (en) | Traffic prediction method of BiLSTM and LightGBM model based on attention mechanism | |
| CN115324843A (en) | Wind generating set fault diagnosis system and method based on monitoring data | |
| CN116691418B (en) | Charging method capable of automatically distributing control power | |
| CN117175588B (en) | Space-time correlation-based electricity load prediction method and device | |
| CN116292130A (en) | Wind driven generator state monitoring system and method based on cooperative analysis | |
| Arunan et al. | A federated learning-based industrial health prognostics for heterogeneous edge devices using matched feature extraction | |
| CN117689188B (en) | Big data-based user charging strategy optimization system and method | |
| Fu et al. | ProgRPGAN: Progressive GAN for route planning | |
| CN113505879B (en) | Prediction method and device based on multi-attention feature memory model | |
| CN113033089B (en) | Method and device for identifying abnormal electricity consumption user | |
| Bosma et al. | Estimating solar and wind power production using computer vision deep learning techniques on weather maps | |
| CN115310355A (en) | Multi-energy coupling-considered multi-load prediction method and system for comprehensive energy system | |
| CN113536508A (en) | Method and system for classifying manufacturing network nodes | |
| CN117833243B (en) | Method, system, equipment and medium for predicting short-term demand of electric power | |
| CN116340767B (en) | Electric automobile travel energy consumption probability distribution prediction method, system and product | |
| CN116188876B (en) | Semi-supervised learning method and semi-supervised learning device based on information mixing | |
| WO2023212390A1 (en) | Neural network methods | |
| KR20220162359A (en) | Artificial intelligence based classification system and method IoT time series data | |
| CN118011220A (en) | Battery pack state of charge estimation method, system and medium | |
| CN116910350A (en) | Model confidence evaluation method, device, equipment and readable storage medium | |
| Sun et al. | Online Traffic Classification Scheme Based on Bidirectional Long-short Term Memory and Attention in Edge Computing Oriented Optical Networks | |
| CN117977028A (en) | Intelligent charging method and system based on vehicle battery monitoring |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |