CN113205280B - Electric vehicle charging station address selection method of plum cluster guiding attention reasoning network - Google Patents
Electric vehicle charging station address selection method of plum cluster guiding attention reasoning network Download PDFInfo
- Publication number
- CN113205280B CN113205280B CN202110590249.2A CN202110590249A CN113205280B CN 113205280 B CN113205280 B CN 113205280B CN 202110590249 A CN202110590249 A CN 202110590249A CN 113205280 B CN113205280 B CN 113205280B
- Authority
- CN
- China
- Prior art keywords
- electric vehicle
- charging station
- attention
- reasoning network
- grid
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/60—Other road transportation technologies with climate change mitigation effect
- Y02T10/70—Energy storage systems for electromobility, e.g. batteries
Abstract
The invention provides an electric vehicle charging station location method of a prune group guiding attention reasoning network. The method combines a stock machine learning method and a guided attention reasoning network method and is used for electric vehicle charging station site selection. Firstly, the guiding and paying attention to the reasoning network method in the method is mainly used for extracting the characteristics of traffic data and judging the flow information of the electric automobile. Secondly, the Leu machine learning method in the method is used as a parameter of a threshold function of the guided attention reasoning network method, so that training efficiency of the guided attention reasoning network method is improved. And finally, meshing the flow information of the electric vehicle to obtain the charging demand of the electric vehicle, correlating the vehicle flow with grids through longitude and latitude grid division, and determining the optimal charging station address at the dense grids.
Description
Technical Field
The invention belongs to the field of electric power systems, and relates to an electric vehicle charging station site selection method based on an artificial intelligence method, which is suitable for electric power systems and smart city charging station site selection.
Background
Under the international trends of the energy storage crisis of fossil and carbon neutralization, automobile manufacturers start to develop electric automobile industry greatly, get rid of dependence on petroleum, hug clean energy and finish the life of the carbon neutralization of society. A new charging station is built in the existing city network, so that the inclusion degree of the city electric vehicle is improved, and the short plates applied to the electric vehicle are reduced.
The charging station site selection based on data driving is an artificial intelligence based technology, and can solve the defect that some traditional mathematical modeling cannot be completed. The information acquisition based on the machine learning method in the artificial intelligence can comprehensively and rapidly lock data information, effectively extract vehicle information, determine the optimal charging station address through grid information evaluation, and enhance the economy and practicability of charging station address selection.
The stock machine learning combines the advantages of manifold learning and the idea of stock, thereby becoming a learning model with innovative features in the machine learning field. The guiding attention reasoning network is an end-to-end picture recognition mode, is a machine learning method and belongs to a standard mode between supervision and non-supervision.
Therefore, the end-to-end training efficiency can be improved by using the Leu machine learning method as a threshold function parameter of the guided attention reasoning network method, and finally, a large amount of actual monitoring images are used for identifying traffic flow effective information, dividing traffic demand positions and grid codes, and finally, selecting the charging station address of the electric vehicle.
Disclosure of Invention
The invention provides an electric vehicle charging station site selection method of a prune group guiding attention reasoning network, which can refine a training chart label, adds an effective mechanism to control network attention in the learning process of machine learning, trains end to end in the learning process, has clear and clear links, can provide self guidance in the training process, and improves the generalization capability for special target tasks. The method combines a stock machine learning method and a guiding attention reasoning network method, is used for electric vehicle charging station site selection, and obtains electric vehicle charging station candidate positions according to traffic data; the method for locating the electric vehicle charging station of the prune group guiding attention reasoning network comprises the following steps in the using process:
step (1): the guided attention reasoning network method is used for processing traffic data, so that the aim of optimizing a training sample set is fulfilled;
for a given image I, in a constraint stream S c,l In, let f l,k Activating the cells k in the first layer, for each class C from the real labels, the gradient of class C with respect to the activation map is s c Neuron importance weights
The gradient for the reflow will be pooled by global averaging:
wherein GAP (·) represents global average pooling;
to integrate all activation maps, a weight matrix w is used c Activating the graph matrix f as a kernel and applying a two-dimensional convolution l Performing a linear rectification function ReLU operation to obtain A c Note that:
A c ＝ReLU(conv(f l ,w c ))
where l is from the last convolutional layer, the features are balanced between detailed spatial information and high-level semantics;
using attention graph a c Generating a soft mask for the original input image to obtain I *c ：
I *c ＝I-(T(A c )⊙I)
In the formula, the element level multiplication, T (A c ) Is a masking function based on threshold operation; to make T (A) c ) Leadership, using sigmoid function as T (A c ) Function: i is a given image;
where ω is a scale parameter, when T (A c ) When greater than sigma, or equal to 0, ensure T (A c ) Approximately equal to 1; note that the loss function of the mining loss aims to minimize class C I *c Is used to determine the predictive score of (a),
wherein s is c (I *c ) I representing class C *c Is a predictive score of (a); n is the true of the imageThe number of real labels;
final self-guiding loss L self Is the classification loss L cl And L am Is the sum of:
L self ＝L cl +αL am
wherein L is cl Is a multi-label soft profit penalty; alpha is a weight parameter, set to 1;
step (2): the Leu machine learning method is used for guiding attention to the masking function parameters of the reasoning network method;
the Leu machine learning method is used for guiding attention to the masking function parameters of the reasoning network method; sigma is the internal mean of the Leu machine learning sample set:
wherein x is ij Represents the jth sample, n, in the ith class in the training sample set of the lie group i Representing the number of training samples in the ith class, and totally comprising C classes;
step (3): applying the prune group guiding attention reasoning network method to electric vehicle charging station site selection;
converting longitude and latitude of a traffic demand departure point and a destination into grid numbers, and indirectly counting traffic demand in grids by counting the number of times of occurrence of the grid numbers; the grid dividing method is to divide an analysis area into grids with equal size according to a certain interval, and the D represents the interval of grid division; when d=0.01°, it means that the analysis area is divided into a plurality of equally spaced grids at intervals of 0.01 ° longitude and latitude, each grid being a rectangular grid of 700m×1000 m;
definition (X) min ,Y min ) For the lower left longitude and latitude coordinates of the analysis area, (X) min ,Y min ) The longitude and latitude coordinates of the upper right corner; (X) i ,Y i ) Is the longitude and latitude coordinates of any one traffic demand point position, then (X i ,Y i ) Location grid number (grid C x ,gridC y ) Is that
gridC x ＝(int)((X i -X min )/W)
gridC y ＝(int)((Y i -Y min )/W)；
(int) is an integer part; and selecting the grid with large demand as a candidate location for locating the electric vehicle charging station.
Compared with the prior art, the invention has the following advantages and effects:
(1) The guided attention reasoning network method is a machine learning method between supervised learning and unsupervised learning, and the end-to-end training in the training set has good recognition effect; and meanwhile, a shielding function is added to prevent excessive training, so that the specific object identified by the image is more stable.
(2) In the masking function, parameters are obtained from a Leu machine learning method, the internal average value of the Leu learning method is used as a threshold matrix, and supervision and non-supervision are directly influenced during balance training, so that the generalization capability of the system is greatly improved.
(3) The method for the plum cluster guided attention reasoning network provided by the invention acts on a ground traffic system, analyzes road surface information, obtains the charging demand of the electric vehicle, associates traffic flow with grids through longitude and latitude grid division, and finally determines the optimal charging station address at the dense grids. The inaccurate and incomplete information acquisition factors are avoided, and the information is directly acquired from the traffic image with higher accuracy.
Drawings
Fig. 1 is a prune group guided attention inference network of the method of the present invention.
Fig. 2 is a data-driven charging station-based site selection method of the present invention.
Detailed Description
The invention provides an electric vehicle charging station site selection method of a plum cluster guided attention reasoning network, which is described in detail below with reference to the accompanying drawings:
fig. 1 is a prune group guided attention inference network of the method of the present invention. Firstly, integrating traffic data pictures into a training set and a testing set, and guiding attention of a training plum cluster to an inference network. After training picture input, stream S is adopted cl The picture is searched, and the area which is helpful for identifying the category is extracted. Obtaining A using ReLU operation c Note that the region I outside the class C of current interest of the network is obtained simultaneously *c . The lie group guided attention inference network attempts to minimize class C I *c Is the most critical masking function T (A c ) And sigma in (2) is the internal mean value of the Leu machine learning sample set.
Fig. 2 is a data-driven charging station-based site selection method of the present invention. The data-driven electric vehicle charging station site selection method is to excavate potential traffic demand distribution positions by analyzing distribution rules of massive mobile position data, and provide data support for electric vehicle charging station site selection. In order to improve the efficiency of data processing, the invention provides a gridding traffic demand statistical method, wherein gridding is also called unitization, is a form of system division and organization, and aims to reduce the complexity of a system by gridding the system, thereby improving the management level and the control effect. The potential traffic demand locations, including departure points and destinations, are extracted using a crowd-sourced attention inference network using a vast amount of mobile location data. Dividing the analysis area into grids with equal interval, associating the traffic demand position with the grids, and counting the traffic demand in each grid. And selecting the grid with large demand as a candidate location for locating the electric vehicle charging station.
Claims (1)
1. The electric vehicle charging station address selection method of the prune group guiding attention reasoning network is characterized by combining a prune group machine learning method and a guiding attention reasoning network method, and is used for electric vehicle charging station address selection, and electric vehicle charging station candidate positions are obtained according to traffic data; the method for locating the electric vehicle charging station of the prune group guiding attention reasoning network comprises the following steps in the using process:
step (1): the guided attention reasoning network method is used for processing traffic data, so that the aim of optimizing a training sample set is fulfilled;
for a given image I, in a constraint stream S c,l In, let f l,k Activating the cells k in the first layer, for each class C from the real labels, the gradient of class C with respect to the activation map is s c Neuron importance weights
The gradient for the reflow will be pooled by global averaging:
wherein GAP (·) represents global average pooling;
to integrate all activation maps, a weight matrix w is used c Activating the graph matrix f as a kernel and applying a two-dimensional convolution l Performing a linear rectification function ReLU operation to obtain A c Note that:
A c ＝ReLU(conv(f l ,w c ))
where l is from the last convolutional layer, the features are balanced between detailed spatial information and high-level semantics;
using attention graph a c Generating a soft mask for the original input image to obtain I *c ：
In the method, in the process of the invention,
representing multiplication at element level, T (A c ) Is a masking function based on threshold operation; to make T (A) c ) Leadership, using sigmoid function as T (A c ) Function: i is a given image;
where ω is a scale parameter, when T (A c ) When greater than sigma, or equal to 0, ensure T (A c ) Approximately equal to 1; note that the loss function of the mining loss aims to minimize class C I *c Is used to determine the predictive score of (a),
wherein s is c (I *c ) I representing class C *c Is a predictive score of (a); n is the number of real labels of the image;
final self-guiding loss L self Is the classification loss L cl And L am Is the sum of:
L self ＝L cl +αL am
wherein L is cl Is a multi-label soft profit penalty; alpha is a weight parameter, set to 1;
step (2): the Leu machine learning method is used for guiding attention to the masking function parameters of the reasoning network method;
the Leu machine learning method is used for guiding attention to the masking function parameters of the reasoning network method; sigma is the internal mean of the Leu machine learning sample set:
wherein x is ij Represents the jth sample, n, in the ith class in the training sample set of the lie group i Representing the number of training samples in the ith class, and totally comprising C classes;
step (3): applying the prune group guiding attention reasoning network method to electric vehicle charging station site selection;
converting longitude and latitude of a traffic demand departure point and a destination into grid numbers, and indirectly counting traffic demand in grids by counting the number of times of occurrence of the grid numbers; the grid dividing method is to divide an analysis area into grids with equal size according to a certain interval, and the D represents the interval of grid division; when d=0.01°, it means that the analysis area is divided into a plurality of equally spaced grids at intervals of 0.01 ° longitude and latitude, each grid being a rectangular grid of 700m×1000 m;
definition (X) min ,Y min ) For the lower left longitude and latitude coordinates of the analysis area, (X) min ,Y min ) The longitude and latitude coordinates of the upper right corner; (X) i ,Y i ) Is the longitude and latitude coordinates of any one traffic demand point position, then (X i ,Y i ) Location grid number (grid C x ,gridC y ) Is that
gridC x ＝(int)((X i -X min )/W)
gridC y ＝(int)((Y i -Y min )/W)；
(int) is an integer part; and selecting the grid with large demand as a candidate location for locating the electric vehicle charging station.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110590249.2A CN113205280B (en) | 2021-05-28 | 2021-05-28 | Electric vehicle charging station address selection method of plum cluster guiding attention reasoning network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110590249.2A CN113205280B (en) | 2021-05-28 | 2021-05-28 | Electric vehicle charging station address selection method of plum cluster guiding attention reasoning network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113205280A CN113205280A (en) | 2021-08-03 |
| CN113205280B true CN113205280B (en) | 2023-06-23 |
Family
ID=77023470
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110590249.2A Active CN113205280B (en) | 2021-05-28 | 2021-05-28 | Electric vehicle charging station address selection method of plum cluster guiding attention reasoning network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113205280B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101017572A (en) * | 2006-02-09 | 2007-08-15 | 三菱电机株式会社 | Computerized method for tracking object in sequence of frames |
| CN110188597A (en) * | 2019-01-04 | 2019-08-30 | 北京大学 | A kind of dense population counting and accurate positioning method and system based on attention mechanism circulation scaling |
| CN111126485A (en) * | 2019-12-24 | 2020-05-08 | 武汉大学 | Lie-KFDA scene classification method and system based on Lie group machine learning kernel function |
| CN111476199A (en) * | 2020-04-26 | 2020-07-31 | 国网湖南省电力有限公司 | Power transmission and transformation project common grave ground identification method based on high-definition aerial survey image |
| CN111680930A (en) * | 2020-06-17 | 2020-09-18 | 云南省设计院集团有限公司 | Electric vehicle charging station site selection evaluation method based on characteristic reachable circle |
| US10873533B1 (en) * | 2019-09-04 | 2020-12-22 | Cisco Technology, Inc. | Traffic class-specific congestion signatures for improving traffic shaping and other network operations |
-
2021
- 2021-05-28 CN CN202110590249.2A patent/CN113205280B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101017572A (en) * | 2006-02-09 | 2007-08-15 | 三菱电机株式会社 | Computerized method for tracking object in sequence of frames |
| CN110188597A (en) * | 2019-01-04 | 2019-08-30 | 北京大学 | A kind of dense population counting and accurate positioning method and system based on attention mechanism circulation scaling |
| US10873533B1 (en) * | 2019-09-04 | 2020-12-22 | Cisco Technology, Inc. | Traffic class-specific congestion signatures for improving traffic shaping and other network operations |
| CN111126485A (en) * | 2019-12-24 | 2020-05-08 | 武汉大学 | Lie-KFDA scene classification method and system based on Lie group machine learning kernel function |
| CN111476199A (en) * | 2020-04-26 | 2020-07-31 | 国网湖南省电力有限公司 | Power transmission and transformation project common grave ground identification method based on high-definition aerial survey image |
| CN111680930A (en) * | 2020-06-17 | 2020-09-18 | 云南省设计院集团有限公司 | Electric vehicle charging station site selection evaluation method based on characteristic reachable circle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113205280A (en) | 2021-08-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110245709B (en) | 3D point cloud data semantic segmentation method based on deep learning and self-attention | |
| CN110889449A (en) | Edge-enhanced multi-scale remote sensing image building semantic feature extraction method | |
| CN111028255B (en) | Farmland area pre-screening method and device based on priori information and deep learning | |
| CN110111345B (en) | Attention network-based 3D point cloud segmentation method | |
| WO2021013190A1 (en) | Meteorological parameter-based high-speed train positioning method and system in navigation blind zone | |
| CN112163367B (en) | Firefly algorithm and cellular automaton fused city expansion simulation prediction method | |
| CN107944472A (en) | A kind of airspace operation situation computational methods based on transfer learning | |
| Tayyebi et al. | A spatial logistic regression model for simulating land use patterns: a case study of the Shiraz Metropolitan area of Iran | |
| CN106485360A (en) | Segmental society's prediction of economic indexes method and system based on overall noctilucence remote sensing | |
| CN117236674B (en) | Urban river network hydrodynamic force accurate regulation and control and water environment lifting method and system | |
| Li et al. | A GCN-based method for extracting power lines and pylons from airborne LiDAR data | |
| CN114444791A (en) | Flood disaster remote sensing monitoring and evaluation method based on machine learning | |
| Parhar et al. | HyperionSolarNet: solar panel detection from aerial images | |
| Xiao et al. | Open-pit mine road extraction from high-resolution remote sensing images using RATT-UNet | |
| CN116307152A (en) | Traffic prediction method for space-time interactive dynamic graph attention network | |
| Su et al. | New particle formation event detection with Mask R-CNN | |
| CN116662468A (en) | Urban functional area identification method and system based on geographic object space mode characteristics | |
| CN113205280B (en) | Electric vehicle charging station address selection method of plum cluster guiding attention reasoning network | |
| CN117151499A (en) | Monitoring and evaluating method and system for homeland space planning | |
| Chen et al. | DGCNN network architecture with densely connected point pairs in multiscale local regions for ALS point cloud classification | |
| CN115983522A (en) | Rural habitat quality evaluation and prediction method | |
| CN115049160A (en) | Method and system for estimating carbon emission of plain industrial city by using space-time big data | |
| Liu et al. | Channel-Spatial attention convolutional neural networks trained with adaptive learning rates for surface damage detection of wind turbine blades | |
| Putri et al. | Dynamic Spatial Modelling of Land Use Change in South Kuta, Bali | |
| CN117458450B (en) | Power data energy consumption prediction analysis method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |