CN109214424A - A method of the new-energy automobile charging time is predicted using regression analysis and clustering method - Google Patents
A method of the new-energy automobile charging time is predicted using regression analysis and clustering method Download PDFInfo
- Publication number
- CN109214424A CN109214424A CN201810874257.8A CN201810874257A CN109214424A CN 109214424 A CN109214424 A CN 109214424A CN 201810874257 A CN201810874257 A CN 201810874257A CN 109214424 A CN109214424 A CN 109214424A
- Authority
- CN
- China
- Prior art keywords
- charging
- cluster
- curve
- new
- complete history
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/3644—Constructional arrangements
- G01R31/3648—Constructional arrangements comprising digital calculation means, e.g. for performing an algorithm
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Pure & Applied Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Mathematical Physics (AREA)
- Evolutionary Biology (AREA)
- Mathematical Optimization (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Evolutionary Computation (AREA)
- Operations Research (AREA)
- Artificial Intelligence (AREA)
- Algebra (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
- Secondary Cells (AREA)
Abstract
The present invention provides a kind of method using regression analysis and clustering method prediction new-energy automobile charging time, belongs to computer application technology.The method of this method regression analysis is fitted historical data, is clustered using k-means clustering algorithm to history charging curve, the charging time of new-energy automobile is predicted by history charging curve.With the development of new-energy automobile, predict battery charge time be it is highly important, this method predicts battery charge time, can greatly promote new-energy automobile user experience by the excavation to charging pile data.
Description
Technical field
The invention belongs to machine learning field, it is related to a kind of filling using regression analysis and clustering method prediction new-energy automobile
The method of electric time.
Background technique
In the modern life, we increasingly be unable to do without this convenient vehicles of automobile, it provides safety to us
Comfortable outside environment greatly improves our life.But car provider just while also bring many ask
Topic, with increasing for automobile quantity, spacious road also becomes crowded in the past, and the pressure of traffic is also increasing, meanwhile,
Automobile is also constantly discharging the pernicious gases such as oxynitrides, carbon monoxide, hydrocarbon, has been further exacerbated by environment
Pollution, exactly under such a background, this newborn vehicles of new-energy automobile progress into our visual field
In.
New-energy automobile is fast-developing as a new industry, new-energy automobile because of its energy-saving and environmental protection, subtract carbon etc.
Characteristic is increasingly becoming the new direction of automobile industry reform.In recent years, the new-energy automobile technology in China was taken its place in the front ranks of the world, and
And the fast-developing stage is come into, therefore the research in new-energy automobile field has become a popular direction.
During universal, there are also the development that some problems hinder new-energy automobile, such as: charging pile problem, automobile batteries problem etc..
The present invention mainly solves the problems, such as the new-energy automobile charging time, and after user starts to charge to automobile, we pass through recurrence
The method of analysis and cluster prediction new-energy automobile is charged to the time of different electricity, provides better service for user, pushes
New-energy automobile field is preferably developed.
Attached drawing 1 is the charging curve of a wherein vehicle, and abscissa is SOC (battery charge percentage), and ordinate is the time
(unit: second).Wherein, it is all integer that the stepped reason of curve, which is that we obtain when obtaining SOC data, so
Different moments it is possible that identical SOC, show curve ladder-like.We can be obtained by analyzing this charging curve
Know, when starting to charge, SOC's vehicle changes with time substantially straight line, when being charged to a certain percentage, due to electricity
The curve of the protection mechanism of pond management software, time-SOC can bend, and charge power can reduce.
Summary of the invention
The present invention by analysis curve feature, propose it is a kind of using regression analysis and clustering method prediction new energy vapour
The method in vehicle charging time, it is therefore intended that the time prediction that new-energy automobile electricity reaches a certain percentage is provided the user with, with
Meets the needs of user is different.
Technical solution of the present invention:
A method of the new-energy automobile charging time is predicted using regression analysis and clustering method, and steps are as follows:
Step 1: historical data learns
(1) completion is carried out to historical data
(1.1) the history charge data of same class charging pile is collected, it is desirable that the initial charge percentage in history charge data
Than drawing charging curve to history charge data between 0%~80%；It to the charging time and is filled using Pearson correlation coefficient
Electric percentage carries out correlation analysis, the definition of Pearson correlation coefficient are as follows:
Wherein, X is the charging time, and Y is charge volume, and Cov (X, Y) is the covariance of X and Y, and Var (X) is the variance of X, Var
(Y) variance for being Y；The absolute value of Pearson correlation coefficient r (X, Y) shows that the charging time more accords with charge percentage closer to 1
Zygonema sexual intercourse；When by analyzing to obtain charge volume below 80% to Pearson correlation coefficient, charging time and charging percentage
Than in a linear relationship；
(1.2) charging curve that each is drawn reversely is mended in 80% part below using linear regression analysis
Entirely, all complete history charging curves are obtained.
(2) complete history charging curve is clustered
For same class charging pile, different initial slopes can be presented in complete history charging curve, close initial slope
Complete history charging curve clusters phenomenon, is gathered using k-means clustering algorithm to all complete history charging curves
Class, the initial slope refer to complete history charging curve in the slope for starting straight line portion；The k-means, which is clustered, to be calculated
Method is as follows:
(2.1) initial slope of every complete history charging curve is extracted；
(2.2) initial slope of K complete history charging curve is randomly selected, and as the first prothyl of each cluster
The heart, total K initial mass centers；
(2.3) the K complete history charging of the initial slope and extraction that calculate separately every complete history charging curve is bent
The distance between line initial slope；
(2.4) every complete history charging curve cluster is formed into K cluster to apart from the smallest one kind；
(2.5) mass center of K cluster is recalculated, and as new cluster standard；Calculate separately again again every it is complete
Whole history charging curve initial slope between the mass center of K cluster at a distance from, every complete history charging curve is clustered again and is arrived
It apart from the smallest one kind, then recalculates the mass center, calculating and distance of K cluster, cluster again, repeatedly, until front and back two
Stop calculating when the mass center of secondary calculating does not change, end of clustering；
(2.6) it assigns K to a different values of n (n >=2), repeats step (2.2) to step (2.5), it is different to obtain n kind
Cluster result, every kind of cluster result include K cluster；
(3) using error sum of squares as objective function, the error for calculating the n kind cluster result that step (2.6) obtain is flat
Fang He:
Wherein, K indicates K cluster in a certain cluster result, ciIndicate the mass center of i-th of cluster in the cluster result,
Dist indicates distance, and x is the initial slope of any one complete history charging curve in cluster result in ith cluster.
Using the different values of K as abscissa, using the SSE being calculated as ordinate, a curve is drawn.With K value
Increase, the SSE being calculated can be smaller and smaller, so obtained curve beginning is the curve of a monotone decreasing.But
Continue to increase with K value, the improvement of SSE will not continue to significantly reduce, i.e. the reduction speed of SSE slows down.According to elbow
Portion's rule, using the value of the K of curve ancon corresponding points as the cluster number of final complete history charging curve, and it is poly- with this
Class number carries out final cluster to complete history charging curve；
(4) method for utilizing Quadratic regression polynomial analysis, the complete history charging curve that step (1.2) is obtained it is curved
Bent portions are fitted, detailed process are as follows:
It (4.1) is bivariate polynomial of order one, One- place 2-th Order multinomial by One- place 2-th Order polynomial transformation are as follows:
Y=aX2+bX+c
Wherein, X is the charging time, and Y is charge volume, is bivariate polynomial of order one by One- place 2-th Order polynomial transformation, even
X2=X1, X=X2, it obtains:
Y=aX1+bX2+c
Then coefficient a, b, c therein are determined using least square method.
(4.2) principle for utilizing least square method obtains each complete history after completion and polynomial regression operation
The matched curve expression formula of charging curve；
(5) m point for taking SOC between 0%~100% every obtained matched curve is calculated this m positions
Time, then the cluster result final according to step (3), calculates average value of all curves at identical SOC in each cluster,
Obtain K charging curve.
Step 2: the prediction in practical charging time
(1) data point is acquired
During a vehicle charges, the battery charge percentage of a new-energy automobile is read at regular intervals
SOC, and record this moment.
(2) it calculates slope, determine grouping
During new-energy automobile charging, after obtaining a new data point every time, new data point and preceding several is utilized
Time data point calculates slope, when the slope that front and back calculates no longer changes, is being charged according to slope determination at this time
Classification belonging to new-energy automobile charging curve.
(3) time of different charge percentages is determined
The classification according to belonging to the new-energy automobile to charge utilizes history charging curve obtained in (5) in step 1
Prediction new-energy automobile reaches the charging time required when the SOC of needs.
Beneficial effects of the present invention: the method for present invention prediction battery charge time different from the past utilizes charging pile
Historical data predicts the new-energy automobile charging time, it is not necessary to further investigate the internal mechanism of battery, also measure electric current, the electricity of battery
The data such as pressure, ohmic internal resistance, data acquisition are easier to, and accuracy is higher.
Detailed description of the invention
It is the wherein corresponding charging curve of a vehicle that Fig. 1 is corresponding；
It is the flow chart of specific implementation process that Fig. 2 is corresponding；
It is the flow chart of historical data study that Fig. 3 is corresponding；
Fig. 4 corresponding is the flow chart predicted in the charging time.
Specific implementation method
Specific method can be divided into historical data study and the charging time predicts two parts, such as Fig. 2.
Step 1: historical data learns, such as Fig. 3
(1) completion is carried out to historical data
(1.1) the history charge data of same class charging pile is collected, it is desirable that the initial charge percentage in history charge data
Than drawing charging curve to history charge data between 0%~80%；It to the charging time and is filled using Pearson correlation coefficient
Electric percentage carries out correlation analysis, when by analyzing to obtain charge volume below 80% to Pearson correlation coefficient, charging time
It is in a linear relationship with charge percentage；
(1.2) charging curve that each is drawn reversely is mended in 80% part below using linear regression analysis
Entirely, all complete history charging curves are obtained.
(2) complete history charging curve is clustered
For same class charging pile, different initial slopes can be presented in complete history charging curve, close initial slope
Complete history charging curve clusters phenomenon, is gathered using k-means clustering algorithm to all complete history charging curves
Class, the initial slope refer to complete history charging curve in the slope for starting straight line portion；The k-means, which is clustered, to be calculated
Method is as follows:
(2.1) initial slope of every complete history charging curve is extracted；
(2.2) initial slope of K=5 complete history charging curve is randomly selected, and as the first of each cluster
The prothyl heart, totally 5 initial mass centers；
(2.3) the 5 complete histories charging for calculating separately the initial slope of every complete history charging curve and extracting is bent
The distance between line initial slope；
(2.4) every complete history charging curve cluster is formed into 5 clusters to apart from the smallest one kind；
(2.5) mass center of 5 clusters is recalculated, and as new cluster standard；Calculate separately again again every it is complete
Whole history charging curve initial slope between the mass center of 5 clusters at a distance from, every complete history charging curve is clustered again and is arrived
It apart from the smallest one kind, then recalculates mass center, calculating and the distance of 5 clusters, cluster again, repeatedly, until front and back two
Stop calculating when the mass center of secondary calculating does not change, end of clustering；
(2.6) it assigns K to n (we take n=5 herein, 6 ..., 25) a different values, repeats step (2.2) to step
(2.5), the different cluster result of n kind is obtained, every kind of cluster result includes K cluster；
(3) using error sum of squares as objective function, it is poly- to calculate n (n=5,6 ..., the 25) kind that step (2.6) obtain
The error sum of squares of class result, using the SSE being calculated as ordinate, draws a song using the different values of K as abscissa
Line.With the increase of K value, the SSE being calculated can be smaller and smaller, so obtained curve beginning is a monotone decreasing
Curve.But continue to increase with K value, the improvement of SSE will not continue to significantly reduce, i.e. the reduction speed of SSE
Slow down.According to ancon rule, using the value of the K of curve ancon corresponding points as the cluster of final complete history charging curve
Number, and final cluster is carried out to complete history charging curve with the cluster number；
(4) method for utilizing Quadratic regression polynomial analysis, the complete history charging curve that step (1.2) is obtained it is curved
Bent portions are fitted, and determine coefficient a, b, c therein.
(4.2) principle for utilizing least square method obtains each complete history after completion and polynomial regression operation
The matched curve expression formula of charging curve；
(5) 101 points for taking SOC between 0%~100% every obtained matched curve are calculated in this 101 points
The time of position, then the cluster result final according to step (3), it is flat at identical SOC to calculate all curves in each cluster
Mean value obtains K charging curve.
Step 2: the prediction in practical charging time, such as Fig. 4
(1) data point is acquired
During a vehicle charges, the battery charge percentage of a new-energy automobile is read at regular intervals
SOC, and record this moment.
(2) it calculates slope, determine grouping
During new-energy automobile charging, after obtaining a new data point every time, new data point and preceding several is utilized
Time data point calculates slope, when the slope that front and back calculates no longer changes, is being charged according to slope determination at this time
Classification belonging to new-energy automobile charging curve.
(3) time of different charge percentages is determined
The classification according to belonging to the new-energy automobile to charge utilizes history charging curve obtained in (5) in step 1
Prediction new-energy automobile reaches the charging time required when the SOC of needs.
During a new-energy automobile charges, we extract the charging of this moment battery at regular intervals
Percentage records the time at this moment, then carries out regression analysis according to the historical data at this moment, calculates it in straight line
Partial slope, until the value of slope is basically unchanged.Judge which group this curve can assign to according to calculated slope value
In, the curve then learnt according to this group of historical data, which can predict this automobile, can reach certain at what moment
One electricity.After each car charges electricity, the charging situation of this vehicle can be learnt again.
Claims (1)
1. a kind of method using regression analysis and clustering method prediction new-energy automobile charging time, which is characterized in that step
It is as follows:
Step 1: historical data learns
(1) completion is carried out to historical data
(1.1) the history charge data of same class charging pile is collected, it is desirable that the initial charge percentage in history charge data exists
Between 0%~80%, charging curve is drawn to history charge data；Using Pearson correlation coefficient to charging time and charging hundred
Divide than carrying out correlation analysis, the definition of Pearson correlation coefficient are as follows:
Wherein, X is the charging time, and Y is charge volume, and Cov (X, Y) is the covariance of X and Y, and Var (X) is the variance of X, Var (Y)
For the variance of Y；The absolute value of Pearson correlation coefficient r (X, Y) shows that the charging time more meets with charge percentage closer to 1
Linear relationship；When by analyzing to obtain charge volume below 80% to Pearson correlation coefficient, charging time and charge percentage
It is in a linear relationship；
(1.2) reversed completion is carried out in 80% part below to the charging curve that each is drawn using linear regression analysis,
Obtain all complete history charging curves；
(2) complete history charging curve is clustered
For same class charging pile, complete history charging curve can be presented different initial slopes, close initial slope it is complete
History charging curve clusters phenomenon, is clustered using k-means clustering algorithm to all complete history charging curves, institute
The initial slope stated refers to complete history charging curve in the slope for starting straight line portion；The k-means clustering algorithm is such as
Under:
(2.1) initial slope of every complete history charging curve is extracted；
(2.2) initial slope of K complete history charging curve is randomly selected, and as the initial mass center of each cluster,
Total K initial mass centers；
(2.3) at the beginning of calculating separately the initial slope of every complete history charging curve and K complete history charging curve of extraction
The distance between beginning slope；
(2.4) every complete history charging curve cluster is formed into K cluster to apart from the smallest one kind；
(2.5) mass center of K cluster is recalculated, and as new cluster standard；Every is calculated separately again again completely to go through
History charging curve initial slope between the mass center of K cluster at a distance from, every complete history charging curve is clustered again to distance
The smallest one kind, then recalculate the mass center, calculating and distance of K cluster, cluster again, repeatedly, until front and back is counted twice
Stop calculating when the mass center of calculation does not change, end of clustering；
(2.6) K is assigned to n different value, n >=2, repeatedly step (2.2) to step (2.5), obtain the different cluster of n kind
As a result, every kind of cluster result includes K cluster；
(3) using error sum of squares as objective function, the error sum of squares for the n kind cluster result that step (2.6) obtain is calculated:
Wherein, K indicates K cluster in a certain cluster result, ciIndicate the mass center of i-th of cluster in the cluster result, dist table
Show distance, x is the initial slope of any one complete history charging curve in cluster result in ith cluster；
Using the different values of K as abscissa, using the SSE being calculated as ordinate, a curve is drawn；With the increase of K value,
The SSE being calculated can be smaller and smaller, so obtained curve beginning is the curve of a monotone decreasing；But with K
Value continues to increase, and the improvement of SSE will not continue to significantly reduce, i.e. the reduction speed of SSE slows down；According to ancon method
Then, using the value of the K of curve ancon corresponding points as the cluster number of final complete history charging curve, and with the cluster
Several pairs of complete history charging curves carry out final cluster；
(4) method for utilizing Quadratic regression polynomial analysis, to the bending section for the complete history charging curve that step (1.2) obtains
Divide and is fitted, detailed process are as follows:
It (4.1) is bivariate polynomial of order one, One- place 2-th Order multinomial by One- place 2-th Order polynomial transformation are as follows:
Y=aX2+bX+c
Wherein, X is the charging time, and Y is charge volume, is bivariate polynomial of order one by One- place 2-th Order polynomial transformation, even X2=
X1, X=X2, it obtains:
Y=aX1+bX2+c
Then coefficient a, b, c therein are determined using least square method；
(4.2) principle for utilizing least square method obtains the charging of each complete history after completion and polynomial regression operation
The matched curve expression formula of curve；
(5) m point for taking SOC between 0%~100% every obtained matched curve, calculate the position this m when
Between, then the cluster result final according to step (3), calculates average value of all curves at identical SOC in each cluster, obtains
K charging curve；
Step 2: the prediction in practical charging time
(1) data point is acquired
During a vehicle charges, the battery charge percentage SOC of a new-energy automobile is read at regular intervals, and
Record this moment；
(2) it calculates slope, determine grouping
During new-energy automobile charging, after obtaining a new data point every time, new data point and preceding several moment are utilized
Data point calculation goes out slope, and when the slope that front and back calculates no longer changes, the new energy to charge is determined according to slope at this time
Classification belonging to the automobile charging curve of source；
(3) time of different charge percentages is determined
The classification according to belonging to the new-energy automobile to charge is predicted using history charging curve obtained in (5) in step 1
New-energy automobile reaches the charging time required when the SOC of needs.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810874257.8A CN109214424B (en) | 2018-08-03 | 2018-08-03 | Method for predicting charging time of new energy automobile by using regression analysis and clustering method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810874257.8A CN109214424B (en) | 2018-08-03 | 2018-08-03 | Method for predicting charging time of new energy automobile by using regression analysis and clustering method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109214424A true CN109214424A (en) | 2019-01-15 |
| CN109214424B CN109214424B (en) | 2020-02-18 |
Family
ID=64988518
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810874257.8A Active CN109214424B (en) | 2018-08-03 | 2018-08-03 | Method for predicting charging time of new energy automobile by using regression analysis and clustering method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109214424B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110674835A (en) * | 2019-03-22 | 2020-01-10 | 集美大学 | Terahertz imaging method and system and nondestructive testing method and system |
| CN112131606A (en) * | 2020-09-24 | 2020-12-25 | 合肥城市云数据中心股份有限公司 | Dynamic data difference privacy histogram publishing method based on K-means + + combined elbow method autonomous clustering technology |
| CN113942415A (en) * | 2021-12-06 | 2022-01-18 | 张志刚 | Electric vehicle charging pile power resource distribution method and system based on big data |
| CN116662629A (en) * | 2023-08-02 | 2023-08-29 | 杭州宇谷科技股份有限公司 | Charging curve retrieval method, system, device and medium based on time sequence clustering |
| CN117406007A (en) * | 2023-12-14 | 2024-01-16 | 山东佰运科技发展有限公司 | Charging pile charging data detection method and system |
| CN116662629B (en) * | 2023-08-02 | 2024-05-28 | 杭州宇谷科技股份有限公司 | Charging curve retrieval method, system, device and medium based on time sequence clustering |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105353312A (en) * | 2015-09-28 | 2016-02-24 | 华晨汽车集团控股有限公司 | Prediction method of power battery SOC |
| CN105631483A (en) * | 2016-03-08 | 2016-06-01 | 国家电网公司 | Method and device for predicting short-term power load |
| CN106156895A (en) * | 2016-07-29 | 2016-11-23 | 国网山东省电力公司经济技术研究院 | A kind of charging electric vehicle load forecasting method based on fuzzy C-means clustering with substep grid search support vector regression |
| CN107145986A (en) * | 2017-05-24 | 2017-09-08 | 北京中电普华信息技术有限公司 | A kind of charge capacity Forecasting Methodology and device |
| US20170286838A1 (en) * | 2016-03-29 | 2017-10-05 | International Business Machines Corporation | Predicting solar power generation using semi-supervised learning |
-
2018
- 2018-08-03 CN CN201810874257.8A patent/CN109214424B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105353312A (en) * | 2015-09-28 | 2016-02-24 | 华晨汽车集团控股有限公司 | Prediction method of power battery SOC |
| CN105631483A (en) * | 2016-03-08 | 2016-06-01 | 国家电网公司 | Method and device for predicting short-term power load |
| US20170286838A1 (en) * | 2016-03-29 | 2017-10-05 | International Business Machines Corporation | Predicting solar power generation using semi-supervised learning |
| CN106156895A (en) * | 2016-07-29 | 2016-11-23 | 国网山东省电力公司经济技术研究院 | A kind of charging electric vehicle load forecasting method based on fuzzy C-means clustering with substep grid search support vector regression |
| CN107145986A (en) * | 2017-05-24 | 2017-09-08 | 北京中电普华信息技术有限公司 | A kind of charge capacity Forecasting Methodology and device |
Non-Patent Citations (4)
| Title |
|---|
| B.S.BINI等: "Clustering and Regression Techniques for Stock Prediction", 《PROCEDIA TECHNOLOGY》 * |
| JINGSHENG LEI等: "Short-term load forecasting with clustering–regression model in distributed cluster", 《CLUSTER COMPUTING》 * |
| MOSTAFA MAJIDPOUR等: "Forecasting the EV charging load based on customer profile or station measurement", 《APPLIED ENERGY》 * |
| 田明光等: "基于K均值聚类及高斯过程回归集成的铅酸电池荷电状态预测", 《设计研究与应用》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110674835A (en) * | 2019-03-22 | 2020-01-10 | 集美大学 | Terahertz imaging method and system and nondestructive testing method and system |
| CN110674835B (en) * | 2019-03-22 | 2021-03-16 | 集美大学 | Terahertz imaging method and system and nondestructive testing method and system |
| CN112131606A (en) * | 2020-09-24 | 2020-12-25 | 合肥城市云数据中心股份有限公司 | Dynamic data difference privacy histogram publishing method based on K-means + + combined elbow method autonomous clustering technology |
| CN113942415A (en) * | 2021-12-06 | 2022-01-18 | 张志刚 | Electric vehicle charging pile power resource distribution method and system based on big data |
| CN113942415B (en) * | 2021-12-06 | 2023-05-05 | 张志刚 | Big data-based electric vehicle charging pile power resource distribution method and system |
| CN116662629A (en) * | 2023-08-02 | 2023-08-29 | 杭州宇谷科技股份有限公司 | Charging curve retrieval method, system, device and medium based on time sequence clustering |
| CN116662629B (en) * | 2023-08-02 | 2024-05-28 | 杭州宇谷科技股份有限公司 | Charging curve retrieval method, system, device and medium based on time sequence clustering |
| CN117406007A (en) * | 2023-12-14 | 2024-01-16 | 山东佰运科技发展有限公司 | Charging pile charging data detection method and system |
| CN117406007B (en) * | 2023-12-14 | 2024-02-13 | 山东佰运科技发展有限公司 | Charging pile charging data detection method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109214424B (en) | 2020-02-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109214424A (en) | A method of the new-energy automobile charging time is predicted using regression analysis and clustering method | |
| CN106203720B (en) | A kind of schedulable capacity prediction methods of Multiple Time Scales electric car cluster | |
| CN107909179B (en) | Method for constructing prediction model of running condition of plug-in hybrid vehicle and vehicle energy management method | |
| CN110775043B (en) | Hybrid electric vehicle energy optimization method based on battery life attenuation pattern recognition | |
| CN104699606A (en) | Method for predicting state of software system based on hidden Markov model | |
| CN107146013A (en) | A kind of classifying type electric automobile demand spatial and temporal distributions dynamic prediction method based on gray prediction and SVMs | |
| Guo et al. | A novel energy consumption prediction model with combination of road information and driving style of BEVs | |
| CN106599615A (en) | Sequence characteristic analysis method for forecasting miRNA target gene | |
| CN108556682A (en) | A kind of continual mileage prediction technique, device and equipment | |
| CN103745110A (en) | Method of estimating operational driving range of all-electric buses | |
| CN112327168A (en) | XGboost-based electric vehicle battery consumption prediction method | |
| CN109738825A (en) | Retired battery active volume and internal resistance detection method and device based on historical data | |
| CN103336998B (en) | A kind of wind energy turbine set fluctuation of power stabilizes the optimized calculation method of target value | |
| Chen et al. | A convolutional neural network-based driving cycle prediction method for plug-in hybrid electric vehicles with bus route | |
| CN103983920A (en) | Method for establishing model of power battery of electric vehicle | |
| CN115759462A (en) | Charging behavior prediction method and device for electric vehicle user and electronic equipment | |
| CN112305441A (en) | Power battery health state assessment method under integrated clustering | |
| CN111291782B (en) | Accumulated load prediction method based on information accumulation k-Shape clustering algorithm | |
| CN111191824A (en) | Power battery capacity attenuation prediction method and system | |
| CN114954022A (en) | Electric automobile cloud cooperative control device and method | |
| CN112036602A (en) | 5G electric vehicle charging prediction method and system integrating human-computer intelligence | |
| Yu et al. | A-EMCS for PHEV based on real-time driving cycle prediction and personalized travel characteristics | |
| CN113715629B (en) | Residual driving range prediction method based on improved symbolic regression and XGBoost algorithm | |
| Chen et al. | A novel method of developing driving cycle for electric vehicles to evaluate the private driving habits | |
| CN108830414B (en) | Load prediction method for commercial charging area of electric automobile |
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 |