CN115691140B - Analysis and prediction method for space-time distribution of automobile charging demand - Google Patents
Analysis and prediction method for space-time distribution of automobile charging demand Download PDFInfo
- Publication number
- CN115691140B CN115691140B CN202211364256.1A CN202211364256A CN115691140B CN 115691140 B CN115691140 B CN 115691140B CN 202211364256 A CN202211364256 A CN 202211364256A CN 115691140 B CN115691140 B CN 115691140B
- Authority
- CN
- China
- Prior art keywords
- charging
- road
- automobile
- clustering
- traffic flow
- 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
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000009826 distribution Methods 0.000 title claims abstract description 34
- 238000004458 analytical method Methods 0.000 title claims abstract description 15
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 24
- 230000008569 process Effects 0.000 claims abstract description 14
- 238000005070 sampling Methods 0.000 claims abstract description 12
- 238000012545 processing Methods 0.000 claims abstract description 11
- 238000004088 simulation Methods 0.000 claims abstract description 9
- 230000008878 coupling Effects 0.000 claims abstract description 7
- 238000010168 coupling process Methods 0.000 claims abstract description 7
- 238000005859 coupling reaction Methods 0.000 claims abstract description 7
- 239000011159 matrix material Substances 0.000 claims description 22
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000000342 Monte Carlo simulation Methods 0.000 claims description 8
- 239000013598 vector Substances 0.000 claims description 5
- 238000012614 Monte-Carlo sampling Methods 0.000 claims description 3
- 230000002776 aggregation Effects 0.000 claims description 3
- 238000004220 aggregation Methods 0.000 claims description 3
- 239000000178 monomer Substances 0.000 claims description 3
- 230000007704 transition Effects 0.000 claims description 3
- 102000003780 Clusterin Human genes 0.000 claims 1
- 108090000197 Clusterin Proteins 0.000 claims 1
- 238000004590 computer program Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 7
- 230000006399 behavior Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000003860 storage Methods 0.000 description 3
- 230000004075 alteration Effects 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 235000012571 Ficus glomerata Nutrition 0.000 description 1
- 244000153665 Ficus glomerata Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Traffic Control Systems (AREA)
Abstract
The application discloses an analysis and prediction method for space-time distribution of automobile charging demands, and belongs to the technical field of operation and planning of electric power systems; the method comprises the following steps: processing an actual data set of the travel and charging of the automobile, and extracting the priority preference of the centralized charging requirement of the automobile by combining a hierarchical clustering algorithm and a K-medoids clustering algorithm; obtaining a correlation coefficient of the preference degree of the charging position and the traffic flow density of the road through the coupling relation of the road network and the power grid; predicting the space-time distribution of the automobile charging demand in the area according to the correlation of the charging place preference priority and the road flow; by analyzing the correlation coefficient of the road traffic density and the user charging preference road section, each vehicle with the charging requirement can be guided into the road section with the highest correlation to be charged in the simulation prediction, and the process of sampling the simulated road distribution is not excessively complicated while the user charging preference characteristic is accurately reflected.
Description
Technical Field
The application belongs to the technical field of operation and planning of power systems, and particularly relates to an analysis and prediction method for space-time distribution of automobile charging demands.
Background
The large-scale access of the electric automobile can have obvious influence on the operation characteristics of the urban power grid, and the urban power grid also faces a new situation of the cooperative coupling operation of the traffic road network and the urban power grid. The electric automobile is influenced by traffic roads, has spatial randomness and uncertainty, and has the intrinsic property of a battery, and the potential charging requirement of the electric automobile can cause overload of a regional power distribution system and local blockage of a line, so that the electric energy quality of a user side is influenced. In order to better improve the power supply quality and the running level of the urban power distribution network, the charging demand space-time characteristics of the electric automobile clusters need to be analyzed and predicted.
The electric vehicle charging demand analysis can be divided into two aspects: (1) Forming a trip probability and a parking charging probability model based on user behaviors for analysis, wherein the research relies on various probability density distribution models and algorithms, and characterization of the actual behaviors of the user may be inaccurate; (2) And analyzing based on the whole travel track data of the electric automobile user, extracting a parking charging preference mode of the user and analyzing the characteristics of the parking charging preference mode by performing cluster analysis on a large amount of real data. At present, most of the charging demand prediction is based on urban functional partitions, and simulation prediction is performed by a Monte Carlo (Monte Carlo) method in consideration of factors such as user route selection, charging behaviors, road traffic conditions and the like, so that the research has a certain significance in an urban planning stage, but the actual traveling behaviors and parking charging demands of users cannot be accurately reflected.
Conventional clustering algorithms currently in common use include a K-means (K-means) algorithm based on location partitioning, a DBSCAN (density-based clustering with noise) algorithm based on density, and a GMM (gaussian mixture model) algorithm based on model. When the algorithms are applied to analyzing the travel full track data of the electric automobile user, some problems still exist: (1) The K-means algorithm is used as an unsupervised clustering algorithm, and is most typically characterized in that the clustering number needs to be given in advance, and the initial clustering number is difficult to give under the conditions of facing massive sample data and having no actual scene reference, so that the stability and accuracy of a clustering effect are affected; (2) In the above clustering algorithm, euclidean distance, mahalanobis distance, etc. are generally used as similarity measurement criteria of the clusters, and using such measurement criteria for GPS coordinate data may ignore conversion errors between different coordinate systems, and generate erroneous clustering results.
Disclosure of Invention
Aiming at the defects of the prior art, the application aims to provide an analysis and prediction method for the space-time distribution of the automobile charging demand, which can accurately reflect the actual traveling behavior and the parking charging demand of a user.
The aim of the application can be achieved by the following technical scheme:
an analysis and prediction method for the space-time distribution of the charging demands of an automobile comprises the following steps:
s1, processing an actual data set of the travel and charging of the automobile, and extracting the priority preference of the centralized charging requirement of the automobile by combining a hierarchical clustering algorithm and a K-medoids (K-central point clustering) clustering algorithm;
s2, obtaining a correlation coefficient of the preference degree of the charging position and the traffic flow density of the road through the coupling relation of the road network and the power grid;
s3, predicting the space-time distribution of the automobile charging demand in the area according to the correlation of the charging place preference priority and the road flow.
Further, in S1, the step of extracting the priority preference of the concentrated charging demand of the automobile includes:
s11, aiming at automobile charging data containing automobile user information, removing data empty records, and extracting effective longitude and latitude coordinate data and charging power data of a charging place;
s12, taking the longitude and latitude coordinate data and the charging power data obtained in the S11 as two independent data sets, inputting an n 1 initialization matrix for hierarchical clustering, and calculating to obtain a distance matrix among each row of vectors by taking the spherical distance as a dissimilarity measure method; determining the optimal clustering center number k according to the maximum inter-class distance in the hierarchical clustering tree to obtain a sample set in the clustering center range;
s13, taking a hierarchical clustering result as the input of K-means clustering, setting an iteration number threshold m, and selecting an initial clustering center Q according to the initial value K of the number of clusters and the sample set of the clustering centers obtained in the step S12; taking each operation result of the K-medoids algorithm as input in the aggregation hierarchy clustering, and returning to the step S12 to obtain the optimal clustering number and the initial clustering center;
s14, outputting a clustering result, calculating entropy of each cluster, and defining the entropy as a preference index of the concentrated charging demands of the automobiles.
Further, in S12, the formula for calculating the distance matrix between the row vectors by using the spherical distance as the dissimilarity measure is as follows:
wherein J represents the object coordinate P i And P j Distance between omega ij Is the object weight, N is the object set, v (P i ,P j ) Representing object P i and Pj The distance within the set of attributes is such that,radian units of latitude and longitude for calculating similarity between two points respectively; r is the earth radius.
Further, in S14, the formula for calculating the entropy of each cluster is:
wherein ,Pij For the cluster point purity in each cluster i, where N i Represents the total number of cluster points in the class, N ij The number of clustering samples in the jth cluster is represented; e, e i The entropy of each cluster, where i is the number of clusters (i= … k).
Further, in the step S2, the step of obtaining the correlation coefficient between the charging position preference degree and the road traffic flow density is:
s21, in each time period T, obtaining daily traffic flow of main traffic intersections of each charging road section through a road traffic platform, calculating a road flow index by adopting a TTI method, establishing a time sequence road traffic flow model, and defining urban road grades according to traffic flow density;
s22, aiming at two random variables, namely road section grade weight and automobile centralized charging demand priority index, defining sample covariance, and judging variable correlation direction by calculating the sample covariance;
s23, obtaining a correlation coefficient of the charging position preference degree and the road traffic flow density by calculating a Spearman grade correlation coefficient;
and S24, determining the priority order of the charging nodes selected in the road traffic network in the travel path of the automobile user according to the correlation coefficient.
Further, in S21, a specific calculation formula of the traffic flow density is as follows:
wherein N is the total road section number of the road traffic topology in the city; v (V) i Is real-timeRoad traffic flow velocity, L i For the length of the traffic section, W i Weighting the road section grade; v (V) free,i Is the free flow speed of the traffic segment.
Further, in S22, the sample variance is defined as:
cov(X,Y)＝E(X)-E(X)E(Y) (6)
further, in S23, a calculation formula of a correlation coefficient between the charging position preference degree and the road traffic flow density is as follows:
in the above formula, n represents the number of sample volumes, and x i ,y i As elements in the matrices X and Y,the sample means for variables X and Y, respectively.
Further, in the step S3, the specific step of predicting the space-time distribution of the vehicle charging demand in the area includes:
s31, establishing a probability distribution model for the starting travel time and the starting place of the automobile user, and establishing an OD matrix to represent all possible paths between the starting place and the stopping place of the automobile travel:
s32, establishing an automobile SOC state model, setting a threshold value for the electric quantity of a battery of the automobile to reach a charging condition, and inputting probability sampling data of Monte Carlo simulation;
s33, randomly extracting departure time, departure place and initial SOC state of a currently simulated vehicle;
s34, dynamically planning a travel path of an automobile user by using a Floyd method with the shortest distance as a target;
s35, performing analog simulation on the position of the current sampling vehicle in the road network and calculating an SOC consumption state;
s36, judging whether the residual electric quantity of the current vehicle in the step S35 is lower than a charging condition threshold value; if yes, obtaining a correlation coefficient of the road traffic flow density and the charging position preference according to the step S23, updating the current position to be a travel starting position, forming a pair of OD matrixes with a travel destination, and modifying an initial planning path to search a proper charging station route for the vehicle again; if not, keeping the original travel route to continue running, and returning to the step S35;
s37, calculating the travel starting time and the SOC state of the vehicle to be charged;
s38, calculating the charging demand of the automobile battery to obtain the charging load of an automobile monomer, and accumulating the charging loads of all Monte Carlo sampling objects to obtain a time sequence load curve of the charging station;
s39, the program judges whether the vehicle is the last sampling vehicle in Monte Carlo simulation; if yes, ending simulation; if not, the process returns to step S33.
Further, the OD matrix is:
wherein n is the total number of traffic road sections in the urban area, and T is the total traffic occurrence of the connecting road sections between any two areas of the city;
the OD matrix can be obtained through the reverse thrust of the traffic flow density of the road, and the specific process is shown in the formula (9):
wherein E is the square sum of estimated errors of traffic flow densities of all urban road sections;all elements to be solved in the OD matrix; i traffic Is road traffic flow; r is the total number of regional road sections; m and n represent the number of user departure points and destination areas, respectively; />Representation for describing road selection process of automobile userState transition probability, which means slave region O i Departure arrival D j The probability of passing the road segment r on the path of (a).
The application has the beneficial effects that:
1) The space-time characteristics of the actual charging demands of the automobile users are analyzed by utilizing a multi-stage clustering method, so that the reduction of clustering effectiveness caused by inaccuracy of manual setting during large-scale sample data set clustering can be avoided, and the subsequent detection workload is greatly reduced;
2) By analyzing the correlation coefficient of the road traffic density and the user charging preference road section, each vehicle with the charging requirement can be guided into the road section with the highest correlation to be charged in the simulation prediction, and the process of sampling the simulated road distribution is not excessively complicated while the user charging preference characteristic is accurately reflected.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described, and it will be obvious to those skilled in the art that other drawings can be obtained according to these drawings without inventive effort.
FIG. 1 is a flow chart of an overall analysis and prediction method of the present application;
FIG. 2 is a flow chart of a multi-stage clustering algorithm of S1 provided by the application;
FIG. 3 is a schematic diagram of a hierarchical clustering tree provided by an embodiment of the present application;
FIG. 4 is a Monte Carlo simulation flow chart of S3 provided by the application;
fig. 5 is a flowchart of the improved Floyd method in S3 provided by the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
As shown in fig. 1, an analysis and prediction method for the space-time distribution of the charging demand of an automobile comprises the following steps:
s1, processing an actual data set of the travel and charging of the automobile, and extracting the priority preference of the concentrated charging requirement of the automobile by using a multi-stage clustering method;
aiming at the problems that a clustering center is difficult to determine, the algorithm complexity is high and the like in the conventional clustering method, the application provides a multi-stage clustering method by combining a hierarchical clustering algorithm and a K-means clustering algorithm, which is used for further obtaining a priority preference index of the centralized charging requirement of an automobile; as shown in fig. 2, the specific steps of extracting the priority preference degree include:
s11, preprocessing data; aiming at a large amount of automobile charging data containing automobile user stay information, checking whether missing values exist or not, removing data empty records, setting screening conditions according to data object types, and extracting effective longitude and latitude coordinate data and charging power data of a charging place.
S12, hierarchical clustering; and (3) taking longitude and latitude coordinate data and charging power data of the charging place obtained by preprocessing the data in the step (S11) as two independent data sets, inputting an n 1 initialization matrix for hierarchical clustering, and calculating to obtain a distance matrix among each row of vectors by taking the spherical distance as a dissimilarity measure method, wherein the calculation formula is as follows:
wherein J represents the object coordinate P i And P j Distance between omega ij Is the object weight, N is the object set, v (P i ,P j ) Representing object P i and Pj The distance in the attribute set is calculated by the following formula:
wherein ,radian units of latitude and longitude for calculating similarity between two points respectively; r is the earth radius, and 6378.1km is taken;
and determining the optimal cluster center number k according to the maximum inter-class distance in the hierarchical cluster tree to obtain a sample set { cluster|cluster (1), …, cluster (k) } in the cluster center range shown in fig. 3.
S13, clustering K-medoids: taking the hierarchical clustering result as the input of K-medoids clustering, setting an iteration number threshold m, and selecting an initial clustering center Q according to the initial value K of the clustering number and the sample set of the clustering centers obtained in the step S12; taking each operation result of the K-medoids algorithm as input in the aggregation hierarchy clustering, and returning to the step S12 to obtain the optimal clustering number and the initial clustering center;
s14, outputting an automobile charging demand priority preference index result, and carrying out visual characterization on an automobile charging demand space-time distribution feature extraction result by combining an urban GIS map;
in the application, the clustering result obtained in the step S13 is externally evaluated through formulas (3) and (4), and the entropy of each cluster is calculated and defined as a priority preference index of the concentrated charging requirement of the automobile:
P ij for the cluster point purity in each cluster i, where N i Represents the total number of cluster points in the class, N ij The number of clustering samples in the jth cluster is represented; e, e i Entropy of each cluster, where i is the number of clusters (i= … k);
the optimal clustering center number and the initial clustering center are determined through the maximum inter-class distance of the sample, and the characteristics of the data are taken as judgment basis, so that the reduction of the clustering effect caused by inaccuracy of manual setting is reduced.
S2, obtaining a correlation coefficient of the preference degree of the charging position and the traffic flow density of the road through the coupling relation of the road network and the power grid; the method comprises the following specific steps of;
s21, in each time period T, obtaining daily traffic flow of main traffic intersections of each charging road section through a road traffic platform, calculating a road flow index by adopting a Travel Time Index (TTI) method, establishing a time sequence road traffic flow model, and defining urban road grades according to traffic flow density;
the specific calculation method of the traffic flow density is as follows:
wherein N is the total road section number of the road traffic topology in the city; v (V) i For real-time road traffic flow velocity, L i For the length of the traffic section, W i Weighting the road section grade; v (V) free,i The free flow speed of the traffic road section;
s22, aiming at two random variables, namely road section grade weight and automobile centralized charging demand priority index, defining a sample covariance as formula (6), and judging the variable correlation direction by calculating the sample covariance;
cov(X,Y)＝E(X)-E(X)E(Y) (6)
s23, summarizing the relation strength of two data samples according to a Spearman (Spearman) correlation coefficient, checking the correlation judgment in the step S22, and analyzing the variable relation between the automobile cluster charging requirement and the road traffic flow; calculating Spearman grade correlation coefficient to obtain correlation coefficient { mu|mu ] of charging position preference degree and road traffic flow density 1 ,μ 2 ,…,μ l The specific calculation process is shown in the formula (7):
in the above formula, n represents the number of sample volumes, and x i ,y i The matrix X and the matrix Y are elements, and X and Y are sample average values of variables X and Y respectively;
and S24, determining the priority order of the charging nodes selected in the road traffic network in the travel path of the automobile user according to the correlation coefficient.
According to the method, the coupling relation between the urban power grid and the traffic road network is considered, the traffic flow density of the road is calculated by adopting a TTI method, and the travel distribution of the urban functional areas in the future is predicted by a growth coefficient method according to the travel current situation of automobile users in the current city and the traffic travel distribution quantity of the urban functional areas; because the growth coefficient method can not reflect factors such as the trip purpose of the automobile user, the road grade and the like among the functional areas, the influence of the urban road grade on the trip distribution of the user is reflected through the TTI road traffic flow density, and the analysis method of the correlation coefficient of the priority preference degree of the charging demand and the traffic flow density is further provided to reflect the influence of the urban road grade on the charging behavior of the automobile user;
in addition, the application also considers the variable relation between the urban road traffic flow and the automobile cluster charging requirement, carries out correlation analysis on the two random variables, and judges the variable correlation direction by calculating the sample covariance. Because the covariance matrix requires that the sample data all conform to a gaussian or gaussian-like distribution, but there may be a nonlinear relationship between the two random variables, and the sample data may not conform to a gaussian or gaussian-like distribution, the application further utilizes Spearman correlation coefficients to test the correlation of the variables and calculate the rank relation coefficients, thereby obtaining a mathematical interpretation of the correlation of the variables.
S3, predicting the space-time distribution of the automobile charging demand in the area according to the correlation between the charging place preference priority and the road flow; the method comprises the following specific steps:
s31, establishing a probability distribution model for the starting travel time and the starting place of the automobile user, and establishing an OD matrix shown in a formula (8) to represent all possible paths between the starting place and the stopping place of the automobile travel:
wherein n is the total number of traffic road sections in the urban area, and T is the total traffic occurrence of the connecting road sections between any two areas of the city;
the OD matrix can be obtained through the reverse thrust of the traffic flow density of the road, and the specific process is shown in the formula (9):
wherein E is the square sum of estimated errors of traffic flow densities of all urban road sections;all elements to be solved in the OD matrix; i traffic Is road traffic flow; r is the total number of regional road sections; m and n represent the number of user departure points and destination areas, respectively;
the discrete markov process is represented by the following conditional probabilities:
wherein Representing the probability of a state transition describing the road selection process for a vehicle user, referred to as the slave region O i Departure arrival D j The probability of passing through the road section r on the path of (a) can be obtained through resident travel investigation statistical data;
s32, establishing an automobile SOC state model, setting a threshold value for the electric quantity of a battery of the automobile to reach a charging condition, and inputting probability sampling data of Monte Carlo simulation;
s33, randomly extracting departure time, departure place and initial SOC state of a currently simulated vehicle as shown in a Monte Carlo simulation flow chart in FIG. 4; the sampling parameters can be obtained by establishing a probability density distribution model through resident travel investigation statistical data;
s34, taking the shortest distance as a target, dynamically planning a travel path of an automobile user by using a Floyd (Floyd) method, wherein a flow chart of the Floyd (Floyd) method is shown in FIG. 5;
s35, performing analog simulation on the position of the current sampling vehicle in the road network and calculating an SOC consumption state;
s36, judging whether the residual electric quantity of the current vehicle in the step S35 is lower than a charging condition threshold value; if yes, obtaining a correlation coefficient of the road traffic flow density and the charging position preference according to the step S23, updating the current position to be a travel starting position, forming a pair of OD matrixes with a travel destination, and modifying an initial planning path to search a proper charging station route for the vehicle again; if not, keeping the original travel route to continue running, and returning to the step S35;
s37, calculating the travel starting time and the SOC state of the vehicle to be charged;
s38, calculating the charging demand of the automobile battery to obtain the charging load of an automobile monomer, and accumulating the charging loads of all Monte Carlo sampling objects to obtain a time sequence load curve of the charging station;
s39, the program judges whether the vehicle is the last sampling vehicle in Monte Carlo simulation; if yes, ending simulation; if not, the process returns to step S33.
An analysis and prediction system for the space-time distribution of the charging demands of an automobile comprises a data processing module, a correlation coefficient calculation module and an analysis and prediction module;
wherein, the data processing module: processing an actual data set of the travel and charging of the automobile, and extracting the priority preference of the centralized charging requirement of the automobile by combining a hierarchical clustering algorithm and a K-medoids clustering algorithm;
and a correlation coefficient calculation module: obtaining a correlation coefficient of the preference degree of the charging position and the traffic flow density of the road through the coupling relation of the road network and the power grid;
an analysis and prediction module; and predicting the space-time distribution of the automobile charging demand in the area according to the correlation coefficient of the charging place preference priority and the road flow.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein. The scheme in the embodiment of the application can be realized by adopting various computer languages, such as object-oriented programming language Java, an transliteration script language JavaScript and the like.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the application.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (4)
1. The analysis and prediction method for the space-time distribution of the automobile charging demand is characterized by comprising the following steps of:
processing an actual data set of the travel and charging of the automobile, and extracting the priority preference of the centralized charging requirement of the automobile by combining a hierarchical clustering algorithm and a K-medoids clustering algorithm;
obtaining a correlation coefficient of the preference degree of the charging position and the traffic flow density of the road through the coupling relation of the road network and the power grid;
predicting the space-time distribution of the automobile charging demand in the area according to the correlation coefficient of the charging place preference priority and the road flow;
the step of extracting the priority preference of the concentrated charging demands of the automobile comprises the following steps:
s11, aiming at automobile charging data containing automobile user information, removing data empty records, and extracting effective longitude and latitude coordinate data and charging power data of a charging place;
s12, inputting longitude and latitude coordinate data and charging power data obtained in the S11 as two independent data sets into hierarchical clusteringInitializing a matrix, and calculating to obtain a distance matrix among the row vectors by taking the spherical distance as a dissimilarity measure method; determining the optimal clustering center number k according to the maximum inter-class distance in the hierarchical clustering tree to obtain a sample set in the clustering center range;
s13, taking hierarchical clustering results as the input of K-medoids clustering, setting an iteration number threshold m, and selecting an initial clustering center according to the initial value K of the clustering number and the sample set of the clustering centers obtained in the step S12The method comprises the steps of carrying out a first treatment on the surface of the Taking each operation result of the K-medoids algorithm as input in the aggregation hierarchy clustering, and returning to the step S12 to obtain the optimal clustering number and the initial clustering center;
s14, outputting a clustering result, calculating entropy of each cluster, and defining the entropy as a preference index of the concentrated charging demands of the automobiles;
in S12, the formula for calculating the distance matrix between each row of vectors by using the spherical distance as the dissimilarity measure is as follows:
(1)
(2)
wherein ,representing object coordinates +.>And->Distance between->Weight for object->For the object set +.>Representation object-> and />Distance within the property set, +.>Radian units of latitude and longitude for calculating similarity between two points respectively; />Is the earth radius;
in S14, the formula for calculating the entropy of each cluster is:
（3）
（4）
wherein ,Pij For each clusterIn (3) cluster point purity, wherein->Representing the total number of cluster points in the class, +.>Indicate->The number of clustering samples in each cluster; e, e i Entropy of each cluster, wherein ∈>Is the number of clusters (+)>)；
The obtaining step of the correlation coefficient of the charging position preference degree and the road traffic flow density comprises the following steps:
s21, in each time period T, obtaining daily traffic flow of main traffic intersections of each charging road section through a road traffic platform, calculating a road flow index by adopting a TTI method, establishing a time sequence road traffic flow model, and defining urban road grades according to traffic flow density;
s22, aiming at two random variables, namely road section grade weight and automobile centralized charging demand priority index, defining sample covariance of the two random variables, and judging variable correlation direction by calculating the sample covariance;
s23, obtaining a correlation coefficient of the charging position preference degree and the road traffic flow density by calculating a Spearman grade correlation coefficient;
s24, determining a priority order of selecting charging nodes in a road traffic network in an automobile user travel path according to the correlation coefficient;
the specific calculation formula of the traffic flow density is as follows:
（5）
wherein ,the total road section number is the total road traffic topology in the city; />For real-time road traffic flow rate->For the length of the traffic section>Weighting the road section grade; />The free flow speed of the traffic road section;
in S23, the calculation formula of the correlation coefficient between the charging position preference degree and the road traffic flow density is as follows:
（7）
in the above-mentioned method, the step of,representing the number of sample volumes, +.>For the elements in matrices X and Y, +.>The sample means for variables X and Y, respectively.
2. The method for analyzing and predicting a space-time distribution of a charging demand of an automobile according to claim 1, wherein in S22, the sample variance is defined as:
（6）。
3. the method for analyzing and predicting the temporal-spatial distribution of the charging demand of an automobile according to claim 1, wherein the specific step of predicting the temporal-spatial distribution of the charging demand of the automobile in the area comprises:
s31, establishing a probability distribution model for the starting travel time and the starting place of the automobile user, and establishing an OD matrix to represent all possible paths between the starting place and the stopping place of the automobile travel:
s32, establishing an automobile SOC state model, setting a threshold value for the electric quantity of a battery of the automobile to reach a charging condition, and inputting probability sampling data of Monte Carlo simulation;
s33, randomly extracting departure time, departure place and initial SOC state of a currently simulated vehicle;
s34, dynamically planning a travel path of an automobile user by using a Floyd method with the shortest distance as a target;
s35, performing analog simulation on the position of the current sampling vehicle in the road network and calculating an SOC consumption state;
s36, judging whether the residual electric quantity of the current vehicle in the step S35 is lower than a charging condition threshold value; if yes, obtaining a correlation coefficient of the road traffic flow density and the charging position preference according to the step S23, updating the current position to be a travel starting position, forming a pair of OD matrixes with a travel destination, and modifying an initial planning path to search a proper charging station route for the vehicle again; if not, keeping the original travel route to continue running, and returning to the step S35;
s37, calculating the travel starting time and the SOC state of the vehicle to be charged;
s38, calculating the charging demand of the automobile battery to obtain the charging load of an automobile monomer, and accumulating the charging loads of all Monte Carlo sampling objects to obtain a time sequence load curve of the charging station;
s39, the program judges whether the vehicle is the last sampling vehicle in Monte Carlo simulation; if yes, ending simulation; if not, the process returns to step S33.
4. The method for analyzing and predicting the space-time distribution of the charging demand of the automobile according to claim 3, wherein the OD matrix is:
（8）
wherein n is the total number of traffic road sections in the urban area, and T is the total traffic occurrence of the connecting road sections between any two areas of the city;
the OD matrix can be obtained through the reverse thrust of the traffic flow density of the road, and the specific process is shown in the formula (9):
（9）
wherein E is the square sum of estimated errors of traffic flow densities of all urban road sections;all elements to be solved in the OD matrix; />Is road traffic flow; />Is the total number of regional road segments; m and->The number of the user departure points and the destination areas are respectively represented; />Representing the probability of a state transition describing the road selection process for the user of the vehicle, from the region +.>Departure arrival->The probability of passing the road segment r on the path of (a).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211364256.1A CN115691140B (en) | 2022-11-02 | 2022-11-02 | Analysis and prediction method for space-time distribution of automobile charging demand |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211364256.1A CN115691140B (en) | 2022-11-02 | 2022-11-02 | Analysis and prediction method for space-time distribution of automobile charging demand |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115691140A CN115691140A (en) | 2023-02-03 |
| CN115691140B true CN115691140B (en) | 2023-08-15 |
Family
ID=85048822
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211364256.1A Active CN115691140B (en) | 2022-11-02 | 2022-11-02 | Analysis and prediction method for space-time distribution of automobile charging demand |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115691140B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115828438B (en) * | 2023-02-20 | 2023-05-05 | 中汽研汽车检验中心（天津）有限公司 | Method, medium and equipment for predicting ultimate performance of automobile |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20160037308A (en) * | 2014-09-26 | 2016-04-06 | 한국전기연구원 | Method for calculating daily load curve of each city based on the place of electric vehicle charging |
| CN107704963A (en) * | 2017-10-13 | 2018-02-16 | 国电南瑞科技股份有限公司 | A kind of charging electric vehicle load forecasting method for considering spatial and temporal distributions |
| CN107945510A (en) * | 2017-11-14 | 2018-04-20 | 东南大学 | A kind of section detecting method for considering transport need and road network operational efficiency |
| CN109447410A (en) * | 2018-09-25 | 2019-03-08 | 华南理工大学 | A kind of charging station site selecting method considering urban road network traffic operating status |
| CN110363332A (en) * | 2019-06-21 | 2019-10-22 | 国网天津市电力公司电力科学研究院 | A kind of electric car charging load spatial and temporal distributions prediction technique based on individual behavior characteristic |
| CN111383452A (en) * | 2019-12-03 | 2020-07-07 | 东南大学 | Method for estimating and predicting short-term traffic running state of urban road network |
| CN112488369A (en) * | 2020-11-19 | 2021-03-12 | 南方电网科学研究院有限责任公司 | Site selection and volume fixing method and device for charging station and storage medium |
| CN112498164A (en) * | 2020-11-30 | 2021-03-16 | 国网北京市电力公司 | Processing method and device of charging strategy |
| CN114676885A (en) * | 2022-03-02 | 2022-06-28 | 三峡大学 | Electric vehicle charging and discharging load space-time distribution prediction method |
-
2022
- 2022-11-02 CN CN202211364256.1A patent/CN115691140B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20160037308A (en) * | 2014-09-26 | 2016-04-06 | 한국전기연구원 | Method for calculating daily load curve of each city based on the place of electric vehicle charging |
| CN107704963A (en) * | 2017-10-13 | 2018-02-16 | 国电南瑞科技股份有限公司 | A kind of charging electric vehicle load forecasting method for considering spatial and temporal distributions |
| CN107945510A (en) * | 2017-11-14 | 2018-04-20 | 东南大学 | A kind of section detecting method for considering transport need and road network operational efficiency |
| CN109447410A (en) * | 2018-09-25 | 2019-03-08 | 华南理工大学 | A kind of charging station site selecting method considering urban road network traffic operating status |
| CN110363332A (en) * | 2019-06-21 | 2019-10-22 | 国网天津市电力公司电力科学研究院 | A kind of electric car charging load spatial and temporal distributions prediction technique based on individual behavior characteristic |
| CN111383452A (en) * | 2019-12-03 | 2020-07-07 | 东南大学 | Method for estimating and predicting short-term traffic running state of urban road network |
| CN112488369A (en) * | 2020-11-19 | 2021-03-12 | 南方电网科学研究院有限责任公司 | Site selection and volume fixing method and device for charging station and storage medium |
| CN112498164A (en) * | 2020-11-30 | 2021-03-16 | 国网北京市电力公司 | Processing method and device of charging strategy |
| CN114676885A (en) * | 2022-03-02 | 2022-06-28 | 三峡大学 | Electric vehicle charging and discharging load space-time distribution prediction method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115691140A (en) | 2023-02-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113704956A (en) | Urban road online microscopic simulation method and system based on digital twin technology | |
| CN111582559B (en) | Arrival time estimation method and device | |
| CN105091889A (en) | Hotspot path determination method and hotspot path determination equipment | |
| CN110909924A (en) | City expansion multi-scenario simulation cellular automata method based on cross entropy optimizer | |
| CN112633591B (en) | Space searching method and device based on deep reinforcement learning | |
| Bai et al. | A forecasting method of forest pests based on the rough set and PSO-BP neural network | |
| CN110598917B (en) | Destination prediction method, system and storage medium based on path track | |
| CN110826244B (en) | Conjugated gradient cellular automaton method for simulating influence of rail transit on urban growth | |
| CN110543978A (en) | Traffic flow data prediction method and device based on wavelet neural network | |
| CN112884014A (en) | Traffic speed short-time prediction method based on road section topological structure classification | |
| CN116187835A (en) | Data-driven-based method and system for estimating theoretical line loss interval of transformer area | |
| CN115691140B (en) | Analysis and prediction method for space-time distribution of automobile charging demand | |
| KR101703972B1 (en) | System and method for predicting groundwater potential area using spatial information | |
| CN113240219A (en) | Land utilization simulation and prediction method | |
| Ranpura et al. | Estimation of vehicle control delay using artificial intelligence techniques for heterogeneous traffic conditions | |
| CN116308579B (en) | Particle swarm-SVM-based engineering cost determination method, system and product | |
| CN116662815A (en) | Training method of time prediction model and related equipment | |
| Li et al. | Route restoration method for sparse taxi GPS trajectory based on Bayesian network | |
| CN116151478A (en) | Short-time traffic flow prediction method, device and medium for improving sparrow search algorithm | |
| CN114463978A (en) | Data monitoring method based on rail transit information processing terminal | |
| Ziat et al. | Learning Embeddings for Completion and Prediction of Relationnal Multivariate Time-Series. | |
| Guangyao et al. | Route Restoration Method for Sparse Taxi GPS trajectory based on Bayesian Network. | |
| CN118245561B (en) | Urban and rural planning and mapping result generation method and system | |
| Bae et al. | Calibration of car-following models using a dual genetic algorithm with central composite design | |
| CN112215520B (en) | Multi-target fusion passing method and device, computer equipment and storage medium |
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 |