CN104022552B - A kind of intelligent detecting method controlled for charging electric vehicle - Google Patents
A kind of intelligent detecting method controlled for charging electric vehicle Download PDFInfo
- Publication number
- CN104022552B CN104022552B CN201410267099.1A CN201410267099A CN104022552B CN 104022552 B CN104022552 B CN 104022552B CN 201410267099 A CN201410267099 A CN 201410267099A CN 104022552 B CN104022552 B CN 104022552B
- Authority
- CN
- China
- Prior art keywords
- model
- information
- background analysis
- battery
- submodule
- 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
Landscapes
- Electric Propulsion And Braking For Vehicles (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
Abstract
The present invention provides a kind of intelligent detecting method controlled for charging electric vehicle, comprises the steps: that A, information gathering submodule gather information from front end charging terminal, forms the record of data stream, and in whole charging process, certain interval of time gathers primary information;B, backstage modeling analysis, form background analysis model;C, access control submodule and send control instruction information to front end charging terminal according to the preset charged mode value that background analysis model obtains;D, intelligent charge control detection module traffic logging based on Real-time Collection, use the method for increment sort decision tree to build Decision-Tree Classifier Model;E, contrast local analytics model and background analysis model;F, the comparing result of step E is transferred to background analysis systems soft ware, and with reference to local analytics model anticipation class object value, optimizes background analysis model.The present invention is while reaching local efficient detection function, and background system can be according to the result received continuous adjusting and optimizing background analysis model.
Description
Technical field
The present invention relates to a kind of intelligent detecting method controlled for charging electric vehicle.
Background technology
Development new energy vehicle is one of Main Means of whole world various countries reply energy crisis and environmental conservation, along with lithium
The performance fast lifting such as the service life of ion battery, energy density, various types of new energy vehicles progress into
In the large-scale Demonstration Application stage, certain form of new energy vehicle comes into commercialization stage.
Current domestic electric automobile charging station from the initial stage Olympic Games, World Expo technical identification stage development to having one
The business promotion of set pattern mould and operation phase, associated core charging device, more replacing equipment and monitoring in this development
System hardware and softwares etc. have been realized in industrialization, disclosure satisfy that from equipment performance index and Product Process degree and reliability
Battery altering and the needs of charging, in terms of the practical situation of current popularization and application, one of subject matter existed is: electrical changing station
Charging load is affected relatively big by the rule that vehicle runs, and major part electric automobile runs and there is the peak period peace peak phase sooner or later,
It is close with the peak period of other conventional loads that the load making electrical changing station presents the load fluctuation time big, peak load appearance,
It is unfavorable for that the shifting peak of urban power distribution network adds paddy, and the access to electrical changing station external power brings certain difficulty, causes simultaneously and changes electricity
The charging electricity price major part stood is peak value electricity price, and the operating cost causing electrical changing station is higher
Cause the reason of problems, predominantly Vehicular charging management strategy simple, less economical.Current actual motion
Electric automobile charging station all uses vehicle to go back to station and i.e. changes electricity, and the charging and conversion electric management strategy charged immediately after changing electricity, not for electricity
Electrical automobile operation characteristic and battery charging cost and electrical changing station part throttle characteristics consider, cause electrical changing station charging cost high, negative
The problems such as lotus fluctuation is big.Actually carry out reducing the charging electricity charge in electrical changing station and improve the part throttle characteristics charging control as target
System strategy is feasible, because there is the moving law at peace peak, peak in vehicle in use, and the charging interval of reserve battery and merit
Rate is to be adjusted within the specific limits, can improve part throttle characteristics and the charging economy of electrical changing station to a great extent.
Summary of the invention
For the shortcoming of prior art, it is an object of the invention to provide a kind of intelligence inspection controlled for charging electric vehicle
Survey method.
To achieve these goals, the invention provides a kind of intelligent detecting method controlled for charging electric vehicle,
It comprises the steps:
A, information gathering: information gathering submodule gathers information from front end charging terminal, form the record of data stream, data stream
Record comprises battery information, information of vehicles and user profile;In whole charging process, certain interval of time collection is once believed
Breath;For every data stream record, there is a timestamp as unique mark, distinguished;
B, background analysis detect, and form background analysis model: by battery information, information of vehicles and user profile by transmission
To background analysis systems soft ware;Background analysis systems soft ware is according to battery information, information of vehicles and user profile, at background analysis
Model is retrieved the preset charged pattern model met;
C, instruction control front end charging: access and control the preset charged mode value that submodule obtains according to background analysis model
Send control instruction information to front end charging terminal;
D, local analytics detect, and form local analytics model: intelligent charge controls detection module number based on Real-time Collection
According to stream record, the method for increment sort decision tree is used to build Decision-Tree Classifier Model;
E, contrast local analytics model and background analysis model;The increment decision model that central processing unit contrast builds in this locality
The anticipation class object value of type and the preset charged mode value of background analysis systems soft ware passback, it is judged that both are the most consistent;
F, optimization background analysis model: the comparing result of step E is transferred to background system software system, and with reference to local
Analyze model anticipation class object value, optimize background analysis model.
Realize the Intelligent charging system of the inventive method and be divided into three parts: (1) front end charging terminal: mended electricity by battery
Device, power supply etc. are mended electricity equipment and are constituted, and wherein include equipment running status information, user profile, information of vehicles, battery information
Deng;(2) intelligent charge controls detection module: be used for connecting front end charging terminal and background analysis systems soft ware, from obtaining front end
Information, and according to background system corresponding charge mode model, forward end passback charging control information;(3) background analysis system
System software: according to existing a large amount of historical datas, analyzes and sets up various charge mode model flexibly.
Wherein, intelligent charge controls detection module is that connecting electric automobile intelligence quickly mends electricity daemon software system and intelligence
Quickly mend the key device of electric terminals, transmitted by effective data and control instruction issues, it is achieved intelligence quickly mends electric terminals
Properly functioning and cooperation with electrical network.Intelligent charge controls to need balancing user charge requirement and charging property simultaneously
Can optimization, take into full account the mutual relation affected between the factor of each parameter and each parameter, reasonable disposition weight.Consider each electricity
The charging interval of electrical automobile, charge volume and the charging scheme of demand elasticity every electric automobile of intellectual analysis, rationally determine preferential
Level and interrupt mechanism, and realize the real-time, interactive with user profile and remote control function.Intelligent charge detection control module bag
Contain:
Information gathering submodule: for gathering relevant information from the charging terminal of front end, including battery information (table one), car
Information (table two), user profile (table three).After the Data Integration of three tables, obtain the data parameters collection of required collection.
Wireless telecommunications submodule: in specific region, transmit data by IEEE802.11 wireless transmission protocol.
Access and control submodule: for according to the secure access authority preset, the information of passback background control system is given front
End terminal.
Cache submodule: the data collected for storage, analyzes the intermediate data of process, the control of backstage passback
Information.
Table one: battery information
Sequence number | Title | Explanation |
1 | Timestamp | Information gathering time point |
2 | Battery ID | The unique identification number ID of battery |
3 | Battery status | Battery status: charge or discharge |
4 | Electricity percentage ratio | Electricity 0-100 contained by battery |
5 | Voltage | Battery voltage value |
6 | Electric current | Cell current value |
7 | Fault message | The preset failure value of information |
8 | High temperature values | The maximum temperature value of battery |
9 | Low-temperature values | The lowest temperature angle value of battery |
10 | Model | The signal type of battery |
11 | Preset charged pattern | The preset charged pattern drawn according to background analysis systematic analysis result |
Table two: information of vehicles
Sequence number | Title | Explanation |
1 | Vehicle-state | Battery status: charge or discharge |
2 | Distance travelled | Electricity 0-100 contained by battery |
3 | Voltage | Magnitude of voltage |
4 | Peak power | The peak power that electric vehicle travels |
5 | Model | Type of vehicle |
6 | Owning user ID | The unique identification number ID of ID |
7 | Use battery ID | The unique identification number ID of battery |
Table three: user profile
Sequence number | Title | Explanation |
1 | ID | The unique identification number ID of ID |
2 | Age | Age of user section |
3 | Occupation | Affiliated preset value |
4 | Hobby | Affiliated preset value |
5 | Driving age | Drive vehicle time |
6 | Sex | User's sex |
In the present invention, using the method for increment sort decision tree to build Decision-Tree Classifier Model, this building process can be, example
As shown in Figure 1.Wherein, (X, y) is data record personnel, battery and information of vehicles combined, and X is to comprise institute
Having the vector of property value, y is corresponding target classification.XiFor an attribute-name, X in record1It it is first attribute-name (such as electricity
Cell voltage), xijFor attribute occurrence, x11For attribute X1First probable value (such as 110V).nminFor preset value: be used for sentencing
The disconnected node split that whether performs assesses (same alike result value belongs to same leaf node minimum number).
Information gain is formula of mathematical.Assume that S is the set of s data sample.Assuming that class label attribute has m
Different value.siBe classification be CiSample number, then entropy or the expectation information of sample set is
Assume that attribute A has v different value { a1,a2,a3,...av,};With attribute A, S can be divided into v subset
{S1,S2,S3,...Sv, wherein SjIt is a for attribute A value in SjSample set;sijIt it is subset SjMiddle classification is CiSample
Number, then be divided into many interval entropys by attribute A or expectation information be
Its middle termFor subset SjPower, equal to subset SjSample number total divided by the sample in S
Number.Given subset Sj,
Wherein,For SjMiddle sample belongs to class CiProbability.The information gain then being carried out dividing by attribute A is
Gain (A)=I (s1,s2,s3,...am)-E(A)。
HB is that Hoeffding retrains ε.Wherein, Hoeffding is a kind of Research of Decision Tree Learning.Specifically, classification is asked
Topic is defined below: a training set such as form (x, y) having N number of sample is given, and y is a discrete class declaration.And x is
One vector having d attribute.Each attribute can be that symbol represents, it is also possible to is numeric representation.Target is intended to build
One model: y=f (x), thus with high accuracy predict y according to x in the future.Such as, x can be the purchase of a client
Record, and y indicates whether to one commodity day record of this client.Or x is a cell phone record, and y represents whether it has
There is fraudulent.Most effective most sorting technique also is exactly Research of Decision Tree Learning.Learner is gone out by Decision Tree Inductive
Model, each node therein comprises the test to an attribute.Every tree all leads to a possible output of this tree
Result.Each leaf contains a prediction class.The preparation method of Y=DT (x) be by sample to be tested from root node always under
Walk to leafy node, test with its specific attribute at each node.Corresponding branch is entered according to test result.One certainly
Plan tree is from root node, by recursively testing node, changes what leaf obtained.Testing attribute for each node
Select, use some heuristic rule to compare all available attributes of this node, therefrom select most suitable one.Classification is certainly
Plan learner such as ID3, C4.5 and CART suppose that all of training set can exist in main memory simultaneously.Therefore they can be learned
The sample number practised receives strict restriction.Decision tree learning person based on hard disk supposes all of sample as SLIQ and SPRINT
All exist on hard disk.Them are learnt by reading repeatedly.When the size abruptly increase of training set, the tree that study is the most complicated
Cost is the highest.If data set has arrived greatly the degree that hard disk cannot accommodate, these methods are by utter failure.Target be intended to into
Mass data collection sets one decision tree learning person of instruction.It requires disposably to learn each sample with extremely short, the constant time.
So can will directly excavate online data resource and go to build the tree of the potential complexity of tool with acceptable calculation cost.Want
On the premise of ensureing to consider that a little subset of all training samples is sufficient for, choose a test for each node and belong to
Property.So, for a given Data-Link, some initial samples are used for testing root node, remaining use next life
Become leaf.And selecting testing attribute for leaf, such recurrence is carried out.Use Hoeffding constraint solve to study carefully competing should choose many
Few sample obtains test and belongs to this difficult problem of victory.If the span of a true value stochastic variable r is R.Assuming that r is had n
Independent observed value, and calculate their meansigma methods.Hoeffding retrains i.e.: for credibility 1-δ, variable r's is true
Value is at leastWherein
This process is a dynamic renewal model process, and its advantage is to need not calculate substantial amounts of historical data, it is also possible to
Model, and model changes along with new data change.
In the present invention, intelligent charge controls detection module decision tree analysis based on high-speed data-flow model, has piecemeal
Process the feature of data stream, it is adaptable in the case of calculating resource-constrained, build decision analytic model;And divide with this decision-making
Analysis model quickly mends the backstage detection assisted verification mechanism of electricity system, analysis model based on expert knowledge library, energy as intelligence
Enough according to newly arrived data stream self-adapting ground replacement analysis model.
In the present invention, intelligent charge controls the detection module incremental Modeling Method by local module, by information gathering, certainly
Dynamic detection, scheduling controlling function combine.Local modeling uses the data of Real-time Collection to draw the control information of anticipation, after
Platform is analyzed systems soft ware use and is obtained preset control information based on historical data analysis model index similarity.In conjunction with contrast in real time
And historical data, the control instruction of charging module is sent to front end charging terminal, according to electricity by intelligent charge detection control module
Pond, vehicle and personal information adjust charge mode;Meanwhile, real-time for this locality analysis result is returned to background analysis system, is used for
Improve the analysis model on backstage.
According to another detailed description of the invention of the present invention, the battery information in traffic logging includes: timestamp, battery ID,
Battery status, electricity percentage ratio, voltage, electric current, fault message, high temperature values, low-temperature values, model, preset charged pattern.
According to another detailed description of the invention of the present invention, the information of vehicles in traffic logging includes: in vehicle-state, traveling
Journey, voltage, peak power, model, owning user ID, use battery ID.
According to another detailed description of the invention of the present invention, the user profile in traffic logging includes: ID, age, duty
Industry, hobby, driving age, sex.
According to another detailed description of the invention of the present invention, in whole charging process, continue on the data stream increment of new collection
Replaceme diiion model.
According to another detailed description of the invention of the present invention, the time interval that adjacent twice gathers information is: the 1-10 second.
According to another detailed description of the invention of the present invention, step B comprises the steps:
B1, by central processing unit, call wireless telecommunications submodule;
Battery ID, vehicle ID and ID are sent in wireless telecommunications submodule by B2, central processing unit;
Battery ID, vehicle ID and ID are transferred to backstage by home control network communication protocol and divide by B3, wireless telecommunications submodule
Analysis system software;
B4, background analysis systems soft ware, according to battery ID, vehicle ID and ID, are retrieved in background analysis model and are met
Preset charged pattern model.
According to another detailed description of the invention of the present invention, step C comprises the steps:
The preset charged mode value that backstage is returned by C1, central processing unit, passes to access and controls submodule.
C2, access control submodule and send control instruction information to front end charging terminal according to preset charged mode value.
According to another detailed description of the invention of the present invention, step E comprises the steps:
The preset charged pattern model that E1, background analysis systems soft ware obtain returns to intelligence by wireless telecommunications submodule
Charging detection control module;
The backstage return data received is passed to central processing unit by E2, wireless telecommunications submodule;
The anticipation class object value of the increment decision model that the contrast of E3, central processing unit builds in this locality and background analysis system
The preset charged mode value of system software passback, it is judged that both are the most consistent, and will determine that result returns to wireless telecommunications submodule.
Compared with prior art, the present invention possesses following beneficial effect:
1, method based on incremental build decision tree is dissolved into local intelligent charging control detection module by the present invention, analyzes
The data stream of real-time update.It is different that the method builds model in traditional decision tree, and traditional method needs all data
Import in data base completely and be analyzed, having new data to add when, need to all data (historical data with newly add
Data) again model.The modeling of increment method is based on caching, and uses dynamic increment modeling method, when new data arrives
The when of reaching, only to newly adding data analysis, and by analysis result and built model integration, reach dynamic modeling function;
2, intelligent charge is utilized to control the increment decision model of detection module this locality structure, to the real-time stream collected
Modeling, after this model anticipation class object value (preset charged pattern), with default the filling of background analysis systems soft ware passback
Power mode value contrasts, and by the difference of both analyses, result is returned to background system.This process reaches local efficiently inspection
While brake, background system can be according to the result received, continuous adjusting and optimizing background analysis model.
The present invention is described in further detail below in conjunction with the accompanying drawings.
Accompanying drawing explanation
Fig. 1 is a kind of flow process using the method for increment sort decision tree to build Decision-Tree Classifier Model;
Fig. 2 is the Intelligent charging system realizing embodiment 1 method;
Fig. 3 shows that intelligent charge controls the job step of detection module.
Detailed description of the invention
Embodiment 1
As shown in Figure 2, it is achieved the Intelligent charging system of the present embodiment method is divided into three parts: (1) front end charging terminal:
Mended by battery electrical equipment, power supply etc. mend electricity equipment constitute, wherein include equipment running status information, user profile, information of vehicles,
Battery information etc.;(2) intelligent charge controls detection module: be used for connecting front end charging terminal and background analysis systems soft ware, from
Obtain the information of front end, and according to background system corresponding charge mode model, forward end passback charging control information;(3) after
Platform analyzes systems soft ware: according to existing a large amount of historical datas, analyzes and sets up various charge mode model flexibly.
Intelligent charge controls detection module and includes: 1, information gathering submodule: for gathering from the charging terminal of front end
Relevant information, including battery information, information of vehicles, user profile;After the Data Integration of three tables, obtain the number of required collection
According to parameter set;2, wireless telecommunications submodule: in specific region, transmit number by IEEE802.11 wireless transmission protocol
According to;3, control submodule is accessed: for according to the secure access authority preset, the information of passback background control system is whole to front end
End;4, cache submodule: the data collected for storage, analyzes the intermediate data of process, the control letter of backstage passback
Breath.
As it is shown on figure 3, the key step (method of the present embodiment) that intelligent charge controls detection module is as follows:
Step 001: information gathering submodule from front end charging terminal gather information, this information in whole charging process not
Disconnected collection, for every data stream record, has a timestamp as unique mark, is distinguished;
Step 002: the data of collection pass through central processing unit, call wireless telecommunications submodule;
Step 003: battery ID, vehicle ID and ID information are sent in communication submodule by central processing unit;
Step 004: after battery ID, vehicle ID and ID information are transferred to by communication submodule by home control network communication protocol
Platform analyzes systems soft ware;
Step 005: data stream based on Real-time Collection, uses the method for increment sort decision tree to build decision tree classification mould
Type;
All related datas are integrated into and with acquisition time stamp are by step 006: by battery ID, vehicle ID and ID
Uniquely identified data flow data, using preset charged pattern in battery information as class object value;In whole charging process, no
The disconnected replaceme diiion model using the new data stream increment gathered;
Step 007: background system is according to battery ID, vehicle ID and ID, and retrieval meets in background analysis model
Preset charged pattern model, returns to intelligent charge detection control module by wireless telecommunications submodule;
Step 008: the backstage return data received is passed to central processing unit by communication submodule;
Step 009: the anticipation class object value of the increment decision model that central processing unit contrast builds in this locality (is preset and filled
Power mode) and the preset charged mode value of background analysis systems soft ware passback, it is judged that both are the most consistent, and will determine that result is returned
Back to wireless telecommunications submodule;
Step 010: comparing result is transferred to background system software system by home control network communication protocol by wireless telecommunications submodule
System, for improving the analysis module of background system;
Step 011: the preset charged mode value that backstage is returned by central processing unit, passes to safe access control submodule;
Step 012: access and control submodule according to preset charged mode value transmission control instruction information to front end charging eventually
End.
Although the present invention is disclosed above with preferred embodiment, but it is not limited to the scope that the present invention implements.Any
The those of ordinary skill in field, without departing from the invention scope of the present invention, when a little improvement can be made, the most every according to this
Bright done equal improvement, should be the scope of the present invention and is contained.
Claims (9)
1. the intelligent detecting method controlled for charging electric vehicle, it comprises the steps:
A, information gathering: information gathering submodule gathers information from front end charging terminal, form traffic logging, described data stream
Record comprises battery information, information of vehicles and user profile;In whole charging process, certain interval of time collection is once believed
Breath;For every data stream record, there is a timestamp as unique mark, distinguished;
B, backstage modeling analysis, form background analysis model: after described battery information, information of vehicles and user profile being transferred to
Platform analyzes systems soft ware;Background analysis systems soft ware is according to battery information, information of vehicles and user profile, at background analysis model
The preset charged pattern model that middle retrieval meets;
C, instruction control front end charging: the preset charged mode value that access control submodule obtains according to background analysis model sends
Control instruction information is to front end charging terminal;
D, local analytics detect, and form local analytics model: intelligent charge controls detection module data based on Real-time Collection stream
Record, uses the method for increment sort decision tree to build Decision-Tree Classifier Model;
E, contrast local analytics model and background analysis model;The Decision-Tree Classifier Model that central processing unit contrast builds in this locality
Anticipation class object value and background analysis systems soft ware passback preset charged mode value, it is judged that both are the most consistent;
F, optimization background analysis model: the comparing result of step E is transferred to background analysis systems soft ware, and with reference to local analytics
Model anticipation class object value, optimizes background analysis model.
Method the most according to claim 1, wherein, the battery information in described traffic logging includes: timestamp, battery
ID, battery status, electricity percentage ratio, voltage, electric current, fault message, high temperature values, low-temperature values, model, preset charged pattern.
Method the most according to claim 1, wherein, the information of vehicles in traffic logging includes: in vehicle-state, traveling
Journey, voltage, peak power, model, owning user ID, use battery ID.
Method the most according to claim 1, wherein, the user profile in traffic logging includes: ID, age, duty
Industry, hobby, driving age, sex.
Method the most according to claim 1, wherein, in whole charging process, continues on the data stream increment of new collection
Replaceme diiion model.
Method the most according to claim 1, wherein, the time interval that adjacent twice gathers information is: the 1-10 second.
Method the most according to claim 1, wherein, described step B comprises the steps:
B1, by central processing unit, call wireless telecommunications submodule;
Battery ID, vehicle ID and ID are sent in wireless telecommunications submodule by B2, central processing unit;
Battery ID, vehicle ID and ID are transferred to background analysis system by home control network communication protocol by B3, wireless telecommunications submodule
System software;
B4, background analysis systems soft ware according to battery ID, vehicle ID and ID, retrieve meet pre-in background analysis model
If charge mode model.
Method the most according to claim 1, wherein, described step C comprises the steps:
The preset charged mode value that backstage is returned by C1, central processing unit, passes to access and controls submodule;
C2, access control submodule and send control instruction information to front end charging terminal according to preset charged mode value.
Method the most according to claim 1, wherein, described step E comprises the steps:
The preset charged pattern model that E1, background analysis systems soft ware obtain returns to intelligent charge by wireless telecommunications submodule
Detection control module;
The backstage return data received is passed to central processing unit by E2, wireless telecommunications submodule;
The anticipation class object value of the Decision-Tree Classifier Model that the contrast of E3, central processing unit builds in this locality and background analysis system
The preset charged mode value of software passback, it is judged that both are the most consistent, and will determine that result returns to wireless telecommunications submodule.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410267099.1A CN104022552B (en) | 2014-06-16 | 2014-06-16 | A kind of intelligent detecting method controlled for charging electric vehicle |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410267099.1A CN104022552B (en) | 2014-06-16 | 2014-06-16 | A kind of intelligent detecting method controlled for charging electric vehicle |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104022552A CN104022552A (en) | 2014-09-03 |
CN104022552B true CN104022552B (en) | 2016-08-31 |
Family
ID=51439176
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410267099.1A Active CN104022552B (en) | 2014-06-16 | 2014-06-16 | A kind of intelligent detecting method controlled for charging electric vehicle |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104022552B (en) |
Families Citing this family (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105656109B (en) * | 2014-11-28 | 2019-02-26 | 战炜 | A kind of charging system and charging method |
CN106779905B (en) * | 2016-11-18 | 2018-07-10 | 合肥工业大学 | Recommend the method and device of battery configuration based on user behavior |
CN106682421A (en) * | 2016-12-28 | 2017-05-17 | 湖南坤宇网络科技有限公司 | Boiler central-air-pipe clogging early-warning method based on decision-making tree system |
CN107037367A (en) * | 2017-01-13 | 2017-08-11 | 孔文欣 | Battery uses the measuring method and measuring system of electricity |
CN109459702B (en) * | 2018-11-09 | 2021-11-12 | 爱驰汽车有限公司 | Battery state analysis method and device based on Internet of vehicles |
CN109521366A (en) * | 2018-11-21 | 2019-03-26 | 爱驰汽车(上海)有限公司 | Electric car changes the performance estimating method and device of power battery in power mode |
CN109948664B (en) * | 2019-02-28 | 2021-05-14 | 深圳智链物联科技有限公司 | Charging mode identification method and device, terminal equipment and storage medium |
CN109884533B (en) * | 2019-03-11 | 2022-09-13 | 北京百度网讯科技有限公司 | Method and device for diagnosing battery fault, equipment and storage medium |
CN111125914B (en) * | 2019-12-25 | 2024-03-22 | 广电计量检测(深圳)有限公司 | Water-cooling charge-discharge linkage method and linkage system |
CN111060765B (en) * | 2019-12-26 | 2022-06-14 | 山东鲁软数字科技有限公司智慧能源分公司 | Detection method, device and system for electric vehicle charging equipment |
CN113580990B (en) * | 2021-06-07 | 2023-04-18 | 广东劲天科技有限公司 | Charging management method, terminal and system |
CN115431811A (en) * | 2022-11-08 | 2022-12-06 | 国网江苏省电力有限公司营销服务中心 | Electric vehicle ordered charging control system and method based on power grid data acquisition network |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101401275A (en) * | 2006-03-06 | 2009-04-01 | Abb研究有限公司 | Charge controller |
CN102447294A (en) * | 2010-10-08 | 2012-05-09 | 台达电子工业股份有限公司 | Vehicle charge system with functions of charge efficiency control and self-adaptive charge service |
CN102498630A (en) * | 2009-09-07 | 2012-06-13 | Abb研究有限公司 | Energy storage systems |
CN103400203A (en) * | 2013-07-19 | 2013-11-20 | 西安交通大学 | Electric vehicle charging station load prediction method based on support vector machine |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2006004986A1 (en) * | 2004-06-29 | 2006-01-12 | Pharmix Corporation | Estimating the accuracy of molecular property models and predictions |
-
2014
- 2014-06-16 CN CN201410267099.1A patent/CN104022552B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101401275A (en) * | 2006-03-06 | 2009-04-01 | Abb研究有限公司 | Charge controller |
CN102498630A (en) * | 2009-09-07 | 2012-06-13 | Abb研究有限公司 | Energy storage systems |
CN102447294A (en) * | 2010-10-08 | 2012-05-09 | 台达电子工业股份有限公司 | Vehicle charge system with functions of charge efficiency control and self-adaptive charge service |
CN103400203A (en) * | 2013-07-19 | 2013-11-20 | 西安交通大学 | Electric vehicle charging station load prediction method based on support vector machine |
Also Published As
Publication number | Publication date |
---|---|
CN104022552A (en) | 2014-09-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104022552B (en) | A kind of intelligent detecting method controlled for charging electric vehicle | |
CN108491969A (en) | Spatial Load Forecasting model building method based on big data | |
CN105868942B (en) | The orderly charging schedule method of electric car | |
CN101728868B (en) | Method for classification and forecast of remote measuring power load patterns | |
CN106022614A (en) | Data mining method of neural network based on nearest neighbor clustering | |
CN111738462B (en) | Fault first-aid repair active service early warning method for electric power metering device | |
CN106650963A (en) | Electric car charging equipment detection and maintenance managing method and device | |
CN103698627A (en) | Transformer fault diagnostic method based on gray fuzzy firefly algorithm optimization | |
CN110310039A (en) | A kind of lithium battery Life cycle on-line optimization monitoring system and monitoring method | |
Lin et al. | Intelligent bus operation optimization by integrating cases and data driven based on business chain and enhanced quantum genetic algorithm | |
CN114707292B (en) | Analysis method for voltage stability of distribution network containing electric automobile | |
CN114498634A (en) | Electric vehicle charging load prediction method based on ammeter data | |
Yang et al. | Electric Vehicle Charging Station Location Decision Analysis for a Two‐Stage Optimization Model Based on Shapley Function | |
Chen et al. | Electric vehicle charging navigation strategy based on data driven and deep reinforcement learning | |
Gu et al. | A Digital Load Forecasting Method Based on Digital Twin and Improved GRU | |
Zhao et al. | Location planning of smart charging station based on fuzzy TOPSIS method | |
Wang | SVR short-term traffic flow forecasting model based on spatial-temporal feature selection | |
Li | Energy consumption prediction of public buildings based on PCA-RF-AdaBoost | |
Jiang | [Retracted] Highway Traffic Flow Prediction Model Construction Based on the Gray Theory and BP Neural Network | |
Keyan et al. | Anomaly detection method of distribution network line loss based on hybrid clustering and LSTM | |
CN104463683A (en) | Long-term load prediction device and method in power grid with multiple sources | |
Guang et al. | Location planning of charging station for electric vehicle based on urban traffic flow | |
CN113822473B (en) | Traction energy consumption reasonable interval prediction method based on multidimensional data | |
Patil et al. | Linear Regression Based Demand Forecast Model in Electric Vehicles-LRDF | |
Wang et al. | A SVR-Based Passenger Route Choice Model in Urban Rail Transit Network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |