CN113821911B - Charging station load prediction method, device and system - Google Patents
Charging station load prediction method, device and system Download PDFInfo
- Publication number
- CN113821911B CN113821911B CN202110976969.2A CN202110976969A CN113821911B CN 113821911 B CN113821911 B CN 113821911B CN 202110976969 A CN202110976969 A CN 202110976969A CN 113821911 B CN113821911 B CN 113821911B
- Authority
- CN
- China
- Prior art keywords
- charging
- fitting
- gaussian
- model
- simulation
- 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 80
- 238000004088 simulation Methods 0.000 claims abstract description 185
- 238000004364 calculation method Methods 0.000 claims abstract description 50
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 26
- 238000000342 Monte Carlo simulation Methods 0.000 claims abstract description 25
- 238000009825 accumulation Methods 0.000 claims abstract description 22
- 238000005315 distribution function Methods 0.000 claims description 39
- 238000013500 data storage Methods 0.000 claims description 6
- 238000012935 Averaging Methods 0.000 abstract description 6
- 238000010586 diagram Methods 0.000 description 8
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 2
- 229910052799 carbon Inorganic materials 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013277 forecasting method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006386 neutralization reaction Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Water Supply & Treatment (AREA)
- Game Theory and Decision Science (AREA)
- Primary Health Care (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a method, a device and a system for predicting charging station load. The device comprises an acquisition statistical unit, a model fitting unit and a simulation calculation unit. The system comprises a charging station load prediction device and a charging station. The method, the device and the system improve the accuracy of the prediction result of the load of the charging station by acquiring a charging behavior data set from actual charging data of the electric vehicle to perform Gaussian fitting, performing simulation calculation and accumulation calculation according to a Monte Carlo simulation algorithm to obtain a simulated daily load curve, and averaging all the predicted daily load curves obtained by accumulation; furthermore, the method, the device and the system for predicting the load of the charging station provided by the invention also improve the fitness of the prediction process and the actual situation by considering the fast charging waiting time and the slow charging waiting time, so that the accuracy of the prediction result of the load of the charging station is further improved.
Description
Technical Field
The invention relates to the field of prediction of charging station loads, in particular to a method, a device and a system for predicting the charging station loads.
Background
With the proposal of the targets of carbon peak reaching and carbon neutralization in China, electric energy substitution is more and more emphasized by power grid enterprises and comprehensive energy service providers, especially in the field of electric automobiles. However, the charging load characteristics of the electric vehicle have randomness and uncertainty, and the charging is intensively connected into a charging station to be charged, so that the impact on the load of a power grid is easily generated, and the safe and stable operation of a power system is influenced. In order to enable the charging station to provide charging service better and improve the operation efficiency of the power grid, it is necessary to predict the load of the charging station and provide a reference basis for subsequent work such as site selection planning, coordination optimization and the like. Therefore, the power enterprises and the comprehensive energy service providers are urgently required to find a quick, efficient and accurate method for predicting the load of the charging station.
In the prior art, the prediction is typically made by: firstly, giving simulation parameters; then determining the charging power of a single vehicle, the initial charging state, and the probability distribution characteristics of the initial charging time and the duration of charging; and finally, simulating the charging load of the electric vehicle charging station in the day by adopting Monte Carlo simulation according to the probability distribution characteristics of all random factors, and giving a final charging load curve and the number of daily average charged vehicles.
However, the prior art still has the following defects: the actual charging data of the electric vehicle is not utilized to extract characteristics, determine an optimal probability model and optimal parameters, and the waiting time of fast charging or slow charging of a user is not considered, which all can cause the problems of low accuracy of a prediction result or large deviation.
Therefore, there is a need for a method, an apparatus and a system for predicting a charging station load, so as to solve the above problems in the prior art.
Disclosure of Invention
In view of the above-mentioned problems, an object of the present invention is to provide a method, an apparatus and a system for predicting a charging station load, so as to improve the accuracy of the prediction result.
The invention provides a method for predicting the load of a charging station, which comprises the following steps: acquiring a charging data set and a preset prediction condition of an electric vehicle in a charging station, and performing data statistics on the charging data set according to statistical requirements to acquire a first charging behavior data set; the prediction conditions comprise statistical requirements, model fitting parameters and simulation parameters; performing Gaussian fitting on each data group in the first charging behavior data group according to the model fitting parameters, and thus fitting to obtain a charging behavior Gaussian model group; each Gaussian model in the charging behavior Gaussian model group corresponds to each data group in the first charging behavior data group one to one; and performing simulation according to the charging behavior Gaussian model and the simulation parameters by a Monte Carlo simulation algorithm, so as to obtain a predicted daily load curve of the charging station load.
In one embodiment, the prediction method further comprises: receiving a charging data group of the electric vehicle in the charging station sent by the intelligent acquisition terminal, and storing the charging data group; the charging data set comprises a first charging data set for each charging pile to supply power each time and the number of vehicles entering the charging station in different time periods; the first charging data set comprises fast charging waiting time, slow charging waiting time, starting charging time, ending charging time, charging station entering time, charging station leaving time, battery capacity and initial state of charge.
In one embodiment, the acquiring a charging data set of an electric vehicle in a charging station and a preset prediction condition, and performing data statistics on the charging data set according to statistical requirements to acquire a first charging behavior data set specifically includes: the method comprises the steps of obtaining a charging data group, preset statistical requirements, preset model fitting parameters and preset simulation parameters of an electric vehicle in a charging station, removing invalid values in the charging data group, and carrying out data statistics on the charging data group with the invalid values removed according to the statistical requirements to obtain a first charging behavior data group.
In one embodiment, the process of gaussian fitting to obtain the gaussian model of the charging behavior specifically includes: acquiring a Gaussian distribution probability density function model, a first fitting segment number, a minimum Gaussian fitting segment number, a maximum Gaussian fitting segment number, a fitting stopping precision threshold value and a fitting stopping precision threshold value from the model fitting parameters; the initial value of the first fitting segment number is the minimum Gaussian fitting segment number; taking the first fitting segment number as the segment number, adopting a nonlinear least square method, and performing Gaussian fitting on one data group in the first charging behavior data group according to a Gaussian distribution probability density function model, so as to calculate and obtain a corresponding first charging Gaussian model, a first fitting precision and a first fitting precision improvement value, and taking the first charging Gaussian model as a first charging Gaussian distribution function; judging whether the first fitting precision is greater than the fitting stopping precision threshold value or not; if the first charging Gaussian distribution function is larger than the second charging Gaussian distribution function, outputting the first charging Gaussian distribution function as a charging behavior Gaussian model; if not, judging whether the first fitting precision improvement value is smaller than the fitting stopping precision threshold value or not; if the first charging Gaussian distribution function is smaller than the first charging Gaussian distribution function, outputting the first charging Gaussian distribution function as a charging behavior Gaussian model; otherwise, taking the next value of the current first fitting segment number as a new first fitting segment number within the preset fitting segment number range; the fitting segment number range is from the minimum Gaussian fitting segment number to the maximum Gaussian fitting segment number; taking the new first fitting segment number as the segment number, and performing Gaussian fitting on the data set according to a Gaussian distribution probability density function model by adopting a nonlinear least square method, so as to calculate and obtain a corresponding second charging Gaussian model, a corresponding second fitting precision and a second fitting precision improvement value; judging whether the second fitting precision is greater than the fitting stopping precision threshold value or not; if so, outputting the second charging Gaussian distribution function as a charging behavior Gaussian model; if not, judging whether the second fitting precision improvement value is smaller than the fitting stopping precision threshold value or not; and if the charging state is smaller than the preset charging state, outputting the second charging Gaussian distribution function as a charging behavior Gaussian model.
In an embodiment, the simulation is performed according to the charging behavior gaussian model group and the simulation parameters by using a monte carlo simulation algorithm, so as to calculate and obtain a predicted daily load curve of the charging station load, specifically: acquiring a set simulation frequency, a set simulation time period and scene characteristic parameters from the simulation parameters, and acquiring a charging start time model from the charging behavior Gaussian model group; according to the set simulation time interval in the day, based on the scene characteristic parameters, the charging starting time model and the initial charge state normal distribution model, performing simulation calculation and accumulation calculation according to a Monte Carlo simulation algorithm to obtain and store a simulation daily load curve, and recording the current simulation times; judging whether the current simulation times are not less than the set simulation times; when the current simulation times are not less than the set simulation times, carrying out average calculation on all the simulation daily load curves so as to obtain a predicted daily load curve; when the current simulation times are smaller than the set simulation times, updating the current simulation times to be first current simulation times, according to the time period in the set simulation day, based on the scene characteristic parameters, the charging starting time model and the initial charge state normal distribution model, performing simulation calculation and accumulation calculation according to a Monte Carlo simulation algorithm to obtain and store a first simulation daily load curve, and recording the first current simulation times; judging whether the first current simulation times are not less than the set simulation times; and when the first current simulation times are not less than the set simulation times, carrying out average calculation on all the first simulation daily load curves so as to obtain a predicted daily load curve.
The invention also provides a prediction device of the charging station load, which comprises an acquisition statistical unit, a model fitting unit and a simulation calculation unit, wherein the acquisition statistical unit is used for acquiring the charging data set and the preset prediction condition of the electric vehicle in the charging station, and performing data statistics on the charging data set according to the statistical requirement to acquire a first charging behavior data set; the prediction conditions comprise statistical requirements, model fitting parameters and simulation parameters; the model fitting unit is used for respectively performing Gaussian fitting on each data group in the first charging behavior data group according to the model fitting parameters so as to obtain a charging behavior Gaussian model group through fitting; wherein each gaussian model in the charging behavior gaussian model group corresponds to each data group in the first charging behavior data group one to one: and the simulation calculation unit is used for carrying out simulation according to the charging behavior Gaussian model and the simulation parameters through a Monte Carlo simulation algorithm so as to obtain a predicted daily load curve of the charging station load.
In one embodiment, the prediction device further comprises a data storage unit, wherein the data storage unit is used for receiving a charging data set of the electric vehicle in the charging station sent by the intelligent acquisition terminal and storing the charging data set; the charging data set comprises a first charging data set for each charging pile to supply power each time and the number of vehicles entering the charging station in different time periods; the first charging data set comprises fast charging waiting time, slow charging waiting time, starting charging time, ending charging time, charging station entering time, charging station leaving time, battery capacity and initial state of charge.
In one embodiment, the obtaining statistical unit is further configured to remove an invalid value in the charging data set.
The invention also provides a system for predicting the load of the charging station, which comprises a device for predicting the load of the charging station and the charging station, wherein the device for predicting the load of the charging station is in communication connection with the charging station, the charging station comprises a plurality of parking spaces, a plurality of charging piles and a plurality of intelligent acquisition terminals, each parking space is provided with one charging pile and one or more intelligent acquisition terminals, and the device for predicting the load of the charging station is used for executing the method for predicting the load of the charging station.
In one embodiment, the charging posts include a fast charging post and a slow charging post.
Compared with the prior art, the embodiment of the invention has the following beneficial effects:
the invention provides a method, a device and a system for predicting a load of a charging station, which are used for acquiring a charging behavior data set from actual charging data of an electric vehicle to perform Gaussian fitting, performing simulation calculation and accumulation calculation according to a Monte Carlo simulation algorithm to obtain a simulated daily load curve, and averaging all the predicted daily load curves obtained by accumulation.
Furthermore, the method, the device and the system for predicting the load of the charging station provided by the invention also improve the fitness of the prediction process and the actual situation by considering the fast charging waiting time and the slow charging waiting time, so that the accuracy of the prediction result of the load of the charging station is further improved.
Drawings
The invention will be further described with reference to the accompanying drawings, in which:
fig. 1 shows a flow chart of an embodiment of a method for predicting a charging station load according to the invention;
FIG. 2 illustrates a reference flow diagram for one embodiment of a process of Gaussian fitting to obtain a Gaussian model of charging behavior;
FIG. 3 illustrates a reference flow diagram for one embodiment of a simulation process;
FIG. 4 shows a flow chart of another embodiment of a method for predicting a charging station load according to the present invention;
fig. 5 is a block diagram showing an embodiment of a charge station load prediction apparatus according to the present invention;
fig. 6 shows a block diagram of an embodiment of a system for predicting the load of a charging station according to the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Detailed description of the preferred embodiment
The embodiment of the invention firstly provides an embodiment of a method for predicting the load of a charging station. Fig. 1 shows a flow chart of an exemplary embodiment of a method for predicting a charging station load according to the present invention. As shown in fig. 1, the method comprises the steps of:
and S1, acquiring a charging data group of the electric automobile in the charging station and a preset prediction condition, and performing data statistics on the charging data group according to the statistical requirement to acquire a first charging behavior data group.
In order to analyze and extract the charging behavior of the user by adopting a high-order Gaussian fitting method in the subsequent process and further predict the load of the charging station by combining the characteristics, firstly, the charging behavior characteristics of the user are acquired, and the user charging behavior characterization can be performed by acquiring data related to the behavior habits of the user in the charging data.
Therefore, a charging data group of each electric vehicle in the charging station is obtained firstly, wherein the charging data group comprises a first charging data group for each charging pile to supply power each time and the number of vehicles entering the charging station in different time periods; the first charging data set comprises quick charging waiting time, slow charging waiting time, starting charging time, ending charging time, charging station entering time, charging station leaving time, battery capacity and initial state of charge. The prediction conditions include statistical requirements, model fitting parameters, and simulation parameters.
Specifically, firstly, a charging data group, preset statistical requirements, preset model fitting parameters and preset simulation parameters of an electric vehicle in a charging station are obtained; and then, removing an invalid value in the charging data group, and performing data statistics on the charging data group with the invalid value removed according to statistical requirements to obtain a first charging behavior data group.
In one embodiment, after acquiring the charging data set, the preset statistical requirements, the preset model fitting parameters and the preset simulation parameters of the electric vehicle in the charging station, the invalid values and the abnormal values should be removed, so as to avoid unnecessary error data interference.
In one embodiment, the statistical requirements include: the method comprises the steps of counting the proportion of the number of vehicles entering the charging station in different time periods in one day to the number of vehicles entering the charging station in one day, counting the number of vehicles entering the charging station and actually entering the charging pile for charging, and the like.
And S2, performing Gaussian fitting on each data group in the first charging behavior data group according to the model fitting parameters, and fitting to obtain a charging behavior Gaussian model group.
And the Gaussian models in the charging behavior Gaussian model group obtained through final fitting correspond to the data groups in the first charging behavior data group one to one respectively.
To illustrate in further detail, for reference, fig. 2 shows a reference flow diagram of one embodiment of a process of gaussian fitting to obtain a gaussian model of charging behavior.
The process of gaussian fitting to obtain the gaussian model of the charging behavior specifically comprises the following steps: acquiring a Gaussian distribution probability density function model, a first fitting segment number, a minimum Gaussian fitting segment number, a maximum Gaussian fitting segment number, a fitting stopping precision threshold value and a fitting stopping precision threshold value from the model fitting parameters; the initial value of the first fitting segment number is the minimum Gaussian fitting segment number; taking the first fitting segment number as the segment number, adopting a nonlinear least square method, and performing Gaussian fitting on one data group in the first charging behavior data group according to a Gaussian distribution probability density function model, so as to calculate and obtain a corresponding first charging Gaussian model, a first fitting precision and a first fitting precision improvement value, and taking the first charging Gaussian model as a first charging Gaussian distribution function; judging whether the first fitting precision is greater than the fitting stopping precision threshold value or not; if so, outputting the first charging Gaussian distribution function as a charging behavior Gaussian model; if not, judging whether the first fitting precision improvement value is smaller than the fitting stopping precision threshold value or not; if the first charging Gaussian distribution function is smaller than the first charging Gaussian distribution function, outputting the first charging Gaussian distribution function as a charging behavior Gaussian model; otherwise, taking the next value of the current first fitting segment number as a new first fitting segment number within the preset fitting segment number range; the fitting segment number range is from the minimum Gaussian fitting segment number to the maximum Gaussian fitting segment number; taking the new first fitting segment number as the segment number, and performing Gaussian fitting on the data set according to a Gaussian distribution probability density function model by adopting a nonlinear least square method, so as to calculate and obtain a corresponding second charging Gaussian model, a corresponding second fitting precision and a second fitting precision improvement value; judging whether the second fitting precision is greater than the fitting stopping precision threshold value or not; if so, outputting the second charging Gaussian distribution function as a charging behavior Gaussian model; if not, judging whether the second fitting precision improvement value is smaller than the fitting stopping precision threshold value or not; and if the charging behavior is smaller than the charging behavior threshold value, outputting the second charging Gaussian distribution function as a corresponding charging behavior Gaussian model.
In one embodiment, the start charge time data set corresponds to a start charge time gaussian model.
In the above process, taking a total of 2 segments of fitted gaussian models as an example, a gaussian fitting process for one data group in the first charging behavior data group is described, in an actual situation, a plurality of segments may need to be fitted for gaussian fitting of one data group, and when a plurality of segments of gaussian models need to be fitted, the above process is repeatedly executed. In addition, there are a plurality of data sets in the first charging behavior data set, and in this step, in order to obtain a charging behavior gaussian model set corresponding to the entire first charging behavior data set, the above-mentioned process should be sequentially performed several times for each data set, respectively, so as to obtain respective charging behavior gaussian models in one-to-one correspondence with the respective data sets in the first charging behavior data set, thereby obtaining a charging behavior gaussian model set corresponding to the first charging behavior data set.
Specifically, the gaussian distribution probability density function model can be represented as:
in the formula, an、bn、cnThe model parameter of the n-th section of the Gaussian distribution probability density function model can be solved by using a nonlinear least square methodn、bn、cnAnd the sum of the Gaussian distribution probability density function models of all the sections obtained by solving is the charging behavior Gaussian model, and the charging behavior Gaussian model can represent the distribution condition of charging requirements in a day of the charging station and fully embody the charging behavior of the user.
In one embodiment, the initial value of the first number of fitted segments is the minimum number of gaussian fitted segments.
In one embodiment, the minimum number of gaussian fitting segments is preset to 1, the maximum number of gaussian fitting segments is preset to 5, the stop fitting accuracy threshold is preset to 99.9%, and the stop fitting accuracy improvement threshold is preset to 0.01%.
In one embodiment, the fitting accuracy may be evaluated by calculating an R-square coefficient, i.e., the R-square coefficient may be used as the first fitting accuracy and the second fitting accuracy, and correspondingly, the fitting accuracy improvement value may be determined by the difference between the current R-square coefficient and the previous R-square coefficient. Specifically, the R-square calculation method is as follows:
in the formula, ωiIs a weight coefficient;
fitting the model to corresponding data;
is the mean value of the original data; y isiIs the original data. Wherein the normal value range of the R-square coefficient (namely the fitting precision) is [0,1 ]]The closer the value is to 1, the higher the accuracy of the fitted model is, and the better the fitting effect is.
And S3, performing simulation according to the charging behavior Gaussian model and the simulation parameters by using a Monte Carlo simulation algorithm, thereby obtaining a predicted daily load curve of the charging station load.
Specifically, a set simulation frequency, a set simulation time period in a day and scene characteristic parameters are obtained from the simulation parameters, and a charging start time gaussian model is obtained from the charging behavior gaussian model group; according to the set time interval in the simulation day, based on the scene characteristic parameters, the starting charging time Gaussian model and the initial charge state normal distribution model, performing simulation calculation and accumulation calculation according to a Monte Carlo simulation algorithm to obtain and store a simulation daily load curve, and recording the current simulation times; judging whether the current simulation times are not less than the set simulation times; when the current simulation times are not less than the set simulation times, carrying out average calculation on all the simulation daily load curves so as to obtain a predicted daily load curve; when the current simulation times are smaller than the set simulation times, updating the current simulation times to be first current simulation times, according to the time period in the set simulation day, based on the scene characteristic parameters, the charging starting time model and the initial charge state normal distribution model, performing simulation calculation and accumulation calculation according to a Monte Carlo simulation algorithm to obtain and store a first simulation daily load curve, and recording the first current simulation times; judging whether the first current simulation times are not less than the set simulation times; and when the first current simulation times are not less than the set simulation times, carrying out average calculation on all the first simulation daily load curves so as to obtain a predicted daily load curve.
For further elaboration, by way of reference, FIG. 3 shows a reference flow diagram of one embodiment of a simulation process.
When analog calculation and accumulation calculation are carried out according to a Monte Carlo analog algorithm, firstly, the charging starting time is extracted from a charging starting time data set according to Gaussian distribution, and the initial charge state is extracted from an initial charge state data set according to normal distribution; then, extracting the fast charging waiting time and the slow charging waiting time from the fast charging waiting time data group and the slow charging waiting time data group according to normal distribution; secondly, simulating the charging states of all the train numbers in one day from the preset initial simulation time according to the extracted starting charging time, initial charging state, quick charging waiting time and slow charging waiting time, determining whether the vehicle is charged and what charging mode is selected, and calculating the charging state and charging completion time of each train number in each set simulation time period; then, respectively acquiring and accumulating the charging load curves of each electric automobile in the charging station within each set simulation day time period according to the charging state and the charging completion time of each train number within each set simulation day time period, so as to obtain the simulation day load curve of the current round of simulation, and recording the current simulation times; if the current simulation times reach the set simulation times, acquiring all simulation daily load curves and averaging the simulation daily load curves, thereby outputting a predicted daily load curve; and if the current simulation times do not reach the set simulation times, entering the next round of simulation.
In one embodiment, the number of simulations is set to 300. The set simulation times can be set by self and are not less than 100 times in principle.
The scene characteristic parameters can effectively reflect the charging behavior of the vehicle owner. In one embodiment, the scene characteristic parameters include number of vehicles present in the day, vehicle battery capacity, slow-charging pile charging power, fast-charging pile charging power, slow-charging pile number, fast-charging pile number, charging efficiency, initial state of charge mean, initial state of charge variance, fast-charging wait time mean, fast-charging wait time variance, slow-charging wait time mean, and slow-charging wait time variance.
The embodiment of the invention provides a charging station load prediction method, which comprises the steps of obtaining a charging behavior data set from actual electric vehicle charging data, carrying out Gaussian fitting, carrying out simulation calculation and accumulation calculation according to a Monte Carlo simulation algorithm to obtain a simulation daily load curve, and averaging all the prediction daily load curves obtained by accumulation.
Detailed description of the invention
Furthermore, another embodiment of a method for predicting a load of a charging station is provided in an embodiment of the present invention. Fig. 2 shows a flow chart of a further exemplary embodiment of a method for predicting a charging station load according to the present invention. As shown in fig. 2, the method comprises the steps of:
a1, receiving the charging data group of the electric vehicle in the charging station sent by the intelligent acquisition terminal, and storing the charging data group.
The charging data set comprises a first charging data set for each charging pile to supply power each time and the number of vehicles entering the charging station in different time periods; the first charging data set comprises fast charging waiting time, slow charging waiting time, starting charging time, ending charging time, charging station entering time, charging station leaving time, battery capacity and initial state of charge.
A2, acquiring a charging data group of the electric automobile in the charging station and a preset prediction condition, and performing data statistics on the charging data group according to statistical requirements to acquire a first charging behavior data group.
In order to analyze and extract the charging behavior of the user by adopting a high-order Gaussian fitting method in the subsequent process and further predict the load of the charging station by combining the characteristics, firstly, the charging behavior characteristics of the user are acquired, and the user charging behavior characterization can be performed by acquiring data related to the behavior habits of the user in the charging data.
Therefore, a charging data group of each electric vehicle in the charging station is obtained firstly, wherein the charging data group comprises a first charging data group for each charging pile to supply power each time and the number of vehicles entering the charging station in different time periods; the first charging data set comprises quick charging waiting time, slow charging waiting time, starting charging time, ending charging time, charging station entering time, charging station leaving time, battery capacity and initial state of charge. The prediction conditions include statistical requirements, model fitting parameters, and simulation parameters.
Specifically, firstly, a charging data group, preset statistical requirements, preset model fitting parameters and preset simulation parameters of an electric vehicle in a charging station are obtained; and then, removing an invalid value in the charging data group, and performing data statistics on the charging data group with the invalid value removed according to statistical requirements to obtain a first charging behavior data group.
In one embodiment, after acquiring the charging data set, the preset statistical requirements, the preset model fitting parameters and the preset simulation parameters of the electric vehicle in the charging station, the invalid values and the abnormal values should be removed, so as to avoid unnecessary error data interference.
In one embodiment, the statistical requirements include: the method comprises the steps of counting the proportion of the number of vehicles entering the charging station in different time periods in one day to the number of vehicles entering the charging station in one day, counting the number of vehicles entering the charging station and actually entering the charging pile for charging, and the like.
And A3, respectively carrying out Gaussian fitting on each data group in the first charging behavior data group according to the model fitting parameters, and thus fitting to obtain a charging behavior Gaussian model group.
And the Gaussian models in the charging behavior Gaussian model group obtained through final fitting correspond to the data groups in the first charging behavior data group one to one respectively.
The process of gaussian fitting to obtain the gaussian model of the charging behavior specifically comprises the following steps: acquiring a Gaussian distribution probability density function model, a first fitting segment number, a minimum Gaussian fitting segment number, a maximum Gaussian fitting segment number, a fitting stopping precision threshold value and a fitting stopping precision threshold value from the model fitting parameters; the initial value of the first fitting segment number is the minimum Gaussian fitting segment number; taking the first fitting segment number as the segment number, adopting a nonlinear least square method, and performing Gaussian fitting on one data group in the first charging behavior data group according to a Gaussian distribution probability density function model, so as to calculate and obtain a corresponding first charging Gaussian model, a first fitting precision and a first fitting precision improvement value, and taking the first charging Gaussian model as a first charging Gaussian distribution function; judging whether the first fitting precision is greater than the fitting stopping precision threshold value or not; if so, outputting the first charging Gaussian distribution function as a charging behavior Gaussian model; if not, judging whether the first fitting precision improvement value is smaller than the fitting stopping precision threshold value or not; if the first charging Gaussian distribution function is smaller than the first charging Gaussian distribution function, outputting the first charging Gaussian distribution function as a charging behavior Gaussian model; otherwise, taking the next value of the current first fitting segment number as a new first fitting segment number within the preset fitting segment number range; the fitting segment number range is from the minimum Gaussian fitting segment number to the maximum Gaussian fitting segment number; taking the new first fitting segment number as the segment number, and performing Gaussian fitting on the data set according to a Gaussian distribution probability density function model by adopting a nonlinear least square method, so as to calculate and obtain a corresponding second charging Gaussian model, a corresponding second fitting precision and a second fitting precision improvement value; judging whether the second fitting precision is greater than the fitting stopping precision threshold value or not; if so, outputting the second charging Gaussian distribution function as a charging behavior Gaussian model; if not, judging whether the second fitting precision improvement value is smaller than the fitting stopping precision threshold value or not; and if the charging behavior is smaller than the charging behavior threshold value, outputting the second charging Gaussian distribution function as a corresponding charging behavior Gaussian model.
In one embodiment, the start charge time data set corresponds to a start charge time gaussian model.
In the above process, taking a total of 2 segments of fitted gaussian models as an example, a gaussian fitting process for one data group in the first charging behavior data group is described, in an actual situation, a plurality of segments may need to be fitted for gaussian fitting of one data group, and when a plurality of segments of gaussian models need to be fitted, the above process is repeatedly executed. In addition, there are a plurality of data sets in the first charging behavior data set, and in this step, in order to obtain a charging behavior gaussian model set corresponding to the entire first charging behavior data set, the above-mentioned process should be sequentially performed several times for each data set, respectively, so as to obtain respective charging behavior gaussian models in one-to-one correspondence with the respective data sets in the first charging behavior data set, thereby obtaining a charging behavior gaussian model set corresponding to the first charging behavior data set.
Specifically, the gaussian distribution probability density function model can be represented as:
in the formula, an、bn、cnThe model parameter of the n-th section of the Gaussian distribution probability density function model can be solved by using a nonlinear least square methodn、bn、cnThe sum of the Gaussian distribution probability density function models of all the sections obtained by solving is the Gaussian model of the charging behavior, and the charging behavior isThe Gaussian model can represent the distribution situation of the charging demand of the charging station in one day, and the charging behavior of the user is fully reflected.
In one embodiment, the initial value of the first number of fitted segments is the minimum number of gaussian fitted segments.
In one embodiment, the minimum number of gaussian fitting segments is preset to 1, the maximum number of gaussian fitting segments is preset to 5, the stop fitting accuracy threshold is preset to 99.9%, and the stop fitting accuracy boost threshold is preset to 0.01%.
In one embodiment, the fitting accuracy may be evaluated by calculating an R-square coefficient, i.e., the R-square coefficient may be used as the first fitting accuracy and the second fitting accuracy, and correspondingly, the fitting accuracy improvement value may be determined by the difference between the current R-square coefficient and the previous R-square coefficient. Specifically, the R-square calculation method is as follows:
in the formula, omegaiIs a weight coefficient;
fitting the model to corresponding data;
is the mean value of the original data; y isiIs the original data. Wherein the normal value range of the R-square coefficient (namely the fitting precision) is [0,1 ]]The closer the value is to 1, the higher the accuracy of the fitted model is, and the better the fitting effect is.
And A4, performing simulation according to the charging behavior Gaussian model and the simulation parameters by using a Monte Carlo simulation algorithm, so as to obtain a predicted daily load curve of the charging station load.
Specifically, a set simulation frequency, a set simulation time period in a day and scene characteristic parameters are obtained from the simulation parameters, and a charging start time gaussian model is obtained from the charging behavior gaussian model group; according to the set time interval in the simulation day, based on the scene characteristic parameters, the starting charging time Gaussian model and the initial charge state normal distribution model, performing simulation calculation and accumulation calculation according to a Monte Carlo simulation algorithm to obtain and store a simulation daily load curve, and recording the current simulation times; judging whether the current simulation times are not less than the set simulation times; when the current simulation times are not less than the set simulation times, carrying out average calculation on all the simulation daily load curves so as to obtain a predicted daily load curve; when the current simulation times are smaller than the set simulation times, updating the current simulation times to be first current simulation times, according to the time period in the set simulation day, based on the scene characteristic parameters, the charging starting time model and the initial charge state normal distribution model, performing simulation calculation and accumulation calculation according to a Monte Carlo simulation algorithm to obtain and store a first simulation daily load curve, and recording the first current simulation times; judging whether the first current simulation times are not less than the set simulation times; and when the first current simulation times are not less than the set simulation times, carrying out average calculation on all the first simulation daily load curves so as to obtain a predicted daily load curve.
When analog calculation and accumulation calculation are carried out according to a Monte Carlo analog algorithm, firstly, the charging starting time is extracted from a charging starting time data set according to Gaussian distribution, and the initial charge state is extracted from an initial charge state data set according to normal distribution; then, extracting the fast charging waiting time and the slow charging waiting time from the fast charging waiting time data group and the slow charging waiting time data group according to normal distribution; secondly, simulating the charging states of all the train numbers in one day from the preset initial simulation time according to the extracted starting charging time, initial charging state, quick charging waiting time and slow charging waiting time, determining whether the vehicle is charged and what charging mode is selected, and calculating the charging state and charging completion time of each train number in each set simulation time period; then, respectively acquiring and accumulating the charging load curves of each electric automobile in the charging station within each set simulation day time period according to the charging state and the charging completion time of each train number within each set simulation day time period, so as to obtain the simulation day load curve of the current round of simulation, and recording the current simulation times; if the current simulation times reach the set simulation times, acquiring all simulation daily load curves and averaging the simulation daily load curves, thereby outputting a predicted daily load curve; and if the current simulation times do not reach the set simulation times, entering the next round of simulation.
In one embodiment, the number of simulations is set to 300. The set simulation times can be set by self and are not less than 100 times in principle.
The scene characteristic parameters can effectively reflect the charging behavior of the vehicle owner. In one embodiment, the scene characteristic parameters include number of vehicles present in the day, vehicle battery capacity, slow-charging pile charging power, fast-charging pile charging power, slow-charging pile number, fast-charging pile number, charging efficiency, initial state of charge mean, initial state of charge variance, fast-charging wait time mean, fast-charging wait time variance, slow-charging wait time mean, and slow-charging wait time variance.
The embodiment of the invention provides a method for predicting the load of a charging station, which comprises the steps of obtaining a charging behavior data set from actual charging data of an electric vehicle, carrying out Gaussian fitting, carrying out simulation calculation and accumulation calculation according to a Monte Carlo simulation algorithm to obtain a simulation daily load curve, and averaging all the prediction daily load curves obtained by accumulation, wherein the accuracy of a prediction result of the load of the charging station is improved by the prediction method, the prediction device and the prediction system; furthermore, the charging station load prediction method provided by the embodiment of the invention improves the conformity between the prediction process and the actual situation by considering the fast charging waiting time and the slow charging waiting time, so that the accuracy of the prediction result of the charging station load is further improved.
Detailed description of the preferred embodiment
Besides the method, the embodiment of the invention also provides an embodiment of a device for predicting the load of the charging station. Fig. 3 is a block diagram showing an embodiment of a charging station load prediction apparatus according to the present invention. As shown in fig. 3, the prediction apparatus 1 includes an acquisition statistic unit 11, a model fitting unit 12, and a simulation calculation unit 13.
The obtaining and counting unit 11 is configured to obtain a charging data set of an electric vehicle in the charging station and a preset prediction condition, and perform data counting on the charging data set according to a counting requirement to obtain a first charging behavior data set. The prediction conditions include statistical requirements, model fitting parameters, and simulation parameters. In one embodiment, the obtaining statistics unit is further configured to remove invalid values in the charging data set.
The model fitting unit 12 is configured to perform gaussian fitting on each data set in the first charging behavior data set according to the model fitting parameters, so as to obtain a charging behavior gaussian model set through fitting. And each Gaussian model in the charging behavior Gaussian model group corresponds to each data group in the first charging behavior data group one to one.
The simulation calculation unit 13 is configured to perform simulation according to the charging behavior gaussian model and the simulation parameters by using a monte carlo simulation algorithm, so as to obtain a predicted daily load curve of the load of the charging station.
In one embodiment, the prediction apparatus 1 further includes a data storage unit, and the data storage unit is configured to receive the charging data set of the electric vehicle in the charging station sent by the intelligent acquisition terminal, and store the charging data set. The charging data set comprises a first charging data set for each charging pile to supply power each time and the number of vehicles entering the charging station in different time periods; the first charging data set comprises fast charging waiting time, slow charging waiting time, starting charging time, ending charging time, charging station entering time, charging station leaving time, battery capacity and initial state of charge.
When load prediction of a charging station is required, the prediction device 1 firstly obtains a charging data set and preset prediction conditions of an electric vehicle in the charging station through the obtaining and counting unit 11, and performs data statistics on the charging data set according to statistical requirements to obtain a first charging behavior data set; then, performing gaussian fitting on each data group in the first charging behavior data group respectively through the model fitting unit 12 according to the model fitting parameters, so as to obtain a charging behavior gaussian model group through fitting; finally, simulation is performed through the simulation calculation unit 13 according to the charging behavior gaussian model and the simulation parameters by using a monte carlo simulation algorithm, so that a predicted daily load curve of the charging station load is obtained.
The embodiment of the invention provides a charging station load forecasting device, which is characterized in that a charging behavior data set is obtained from actual electric vehicle charging data to perform Gaussian fitting, simulation calculation and accumulation calculation are performed according to a Monte Carlo simulation algorithm to obtain a simulated daily load curve, all the predicted daily load curves obtained by accumulation are averaged, and the forecasting method, the forecasting device and the forecasting system improve the accuracy of a forecasting result of the charging station load; furthermore, the charging station load prediction device provided by the embodiment of the invention also improves the adaptability of the prediction process and the actual situation by considering the fast charging waiting time and the slow charging waiting time, so that the accuracy of the prediction result of the charging station load is further improved.
Detailed description of the invention
In addition to the above method and apparatus, an embodiment of a system for predicting a load of a charging station is provided in the embodiments of the present invention. Fig. 4 shows a block diagram of an embodiment of a system for predicting the load of a charging station according to the invention. As shown in fig. 4, the prediction system includes a prediction device 1 of charging station load and a charging station 2, the prediction device 1 is communicatively connected to the charging station 2, and the charging station 2 includes a plurality of parking spaces, a plurality of charging piles, and a plurality of intelligent acquisition terminals.
In one embodiment, each parking space is provided with a charging pile and one or more intelligent acquisition terminals, and the prediction device 1 is configured to execute the prediction method of the charging station load as described above.
In one embodiment, the charging posts include a fast charging post and a slow charging post.
When the load of the charging station needs to be predicted, firstly, a charging data set in the charging station 2 is acquired through one or more intelligent acquisition terminals, the charging data set is sent to the prediction device 1, and after the prediction device 1 receives the charging data set, the prediction method of the load of the charging station is executed, so that a predicted daily load curve is finally obtained.
The embodiment of the invention provides a charging station load prediction system, which comprises a charging behavior data set obtained from actual electric vehicle charging data, a simulation calculation and an accumulation calculation are carried out according to a Monte Carlo simulation algorithm to obtain a simulation daily load curve, and all the prediction daily load curves obtained by accumulation are averaged, wherein the prediction method, the prediction device and the prediction system improve the accuracy of the prediction result of the charging station load; furthermore, the charging station load prediction system provided by the embodiment of the invention also improves the conformance of the prediction process and the actual situation by considering the fast charging waiting time and the slow charging waiting time, so that the accuracy of the prediction result of the charging station load is further improved.
The above-mentioned embodiments are provided to further explain the objects, technical solutions and advantages of the present invention in detail, and it should be understood that the above-mentioned embodiments are only examples of the present invention and are not intended to limit the scope of the present invention. It should be understood that any modifications, equivalents, improvements and the like, which come within the spirit and principle of the invention, may occur to those skilled in the art and are intended to be included within the scope of the invention.
Claims (9)
1. A method for predicting a load of a charging station, the method comprising:
acquiring a charging data set and a preset prediction condition of an electric vehicle in a charging station, and performing data statistics on the charging data set according to statistical requirements to acquire a first charging behavior data set; the prediction conditions comprise statistical requirements, model fitting parameters and simulation parameters;
performing Gaussian fitting on each data group in the first charging behavior data group according to the model fitting parameters, and thus fitting to obtain a charging behavior Gaussian model group; each Gaussian model in the charging behavior Gaussian model group corresponds to each data group in the first charging behavior data group one to one; the method includes the steps of performing gaussian fitting on each data set in a first charging behavior data set according to the model fitting parameters, so as to obtain a charging behavior gaussian model set through fitting, and specifically includes: acquiring a Gaussian distribution probability density function model, a first fitting segment number, a minimum Gaussian fitting segment number, a maximum Gaussian fitting segment number, a fitting stopping precision threshold value and a fitting stopping precision threshold value from the model fitting parameters; the initial value of the first fitting segment number is the minimum Gaussian fitting segment number; taking the first fitting segment number as the segment number, adopting a nonlinear least square method, and performing Gaussian fitting on one data group in the first charging behavior data group according to a Gaussian distribution probability density function model, so as to calculate and obtain a corresponding first charging Gaussian model, a first fitting precision and a first fitting precision improvement value, and taking the first charging Gaussian model as a first charging Gaussian distribution function; judging whether the first fitting precision is greater than the fitting stopping precision threshold value or not; if the first charging Gaussian distribution function is larger than the second charging Gaussian distribution function, outputting the first charging Gaussian distribution function as a charging behavior Gaussian model; if not, judging whether the first fitting precision improvement value is smaller than the fitting stopping precision threshold value or not; if the first charging Gaussian distribution function is smaller than the first charging Gaussian distribution function, outputting the first charging Gaussian distribution function as a charging behavior Gaussian model; otherwise, taking the next value of the current first fitting segment number as a new first fitting segment number within the preset fitting segment number range; the fitting segment number range is from the minimum Gaussian fitting segment number to the maximum Gaussian fitting segment number; taking the new first fitting segment number as the segment number, and performing Gaussian fitting on the data set according to a Gaussian distribution probability density function model by adopting a nonlinear least square method, so as to calculate and obtain a corresponding second charging Gaussian model, a corresponding second fitting precision and a second fitting precision improvement value; judging whether the second fitting precision is greater than the fitting stopping precision threshold value or not; if so, outputting the second charging Gaussian distribution function as a charging behavior Gaussian model; if not, judging whether the second fitting precision improvement value is smaller than the fitting stopping precision threshold value or not; if the charging behavior is smaller than the first charging behavior, outputting the second charging Gaussian distribution function as a charging behavior Gaussian model;
and performing simulation according to the charging behavior Gaussian model and the simulation parameters by a Monte Carlo simulation algorithm, so as to obtain a predicted daily load curve of the charging station load.
2. The method of predicting the load of the charging station according to claim 1, wherein the method further comprises:
receiving a charging data group of the electric vehicle in the charging station sent by the intelligent acquisition terminal, and storing the charging data group; the charging data set comprises a first charging data set for each charging pile to supply power each time and the number of vehicles entering the charging station in different time periods; the first charging data set comprises fast charging waiting time, slow charging waiting time, starting charging time, ending charging time, charging station entering time, charging station leaving time, battery capacity and initial state of charge.
3. The method according to claim 2, wherein the acquiring a charging data set of an electric vehicle in the charging station and a predetermined prediction condition, and performing data statistics on the charging data set according to statistical requirements to acquire a first charging behavior data set specifically comprises:
the method comprises the steps of obtaining a charging data set, preset statistical requirements, preset model fitting parameters and preset simulation parameters of the electric vehicle in a charging station, removing invalid values in the charging data set, and carrying out data statistics on the charging data set after the invalid values are removed according to the statistical requirements to obtain a first charging behavior data set.
4. The method for predicting the load of the charging station according to any one of claims 1 to 3, wherein the simulation is performed according to the charging behavior Gaussian model set and the simulation parameters by using a Monte Carlo simulation algorithm, so as to calculate and obtain a predicted daily load curve of the load of the charging station, specifically:
acquiring set simulation times, set simulation time interval in day and scene characteristic parameters from the simulation parameters, and acquiring a charging starting time model from the charging behavior Gaussian model group;
according to the set simulation time interval in the day, based on the scene characteristic parameters, the charging starting time model and the initial charge state normal distribution model, performing simulation calculation and accumulation calculation according to a Monte Carlo simulation algorithm to obtain and store a simulation daily load curve, and recording the current simulation times;
judging whether the current simulation times are not less than the set simulation times;
when the current simulation times are not less than the set simulation times, carrying out average calculation on all the simulation daily load curves so as to obtain a predicted daily load curve;
when the current simulation times are smaller than the set simulation times, updating the current simulation times to be first current simulation times, according to the time period in the set simulation day, based on the scene characteristic parameters, the charging starting time model and the initial charge state normal distribution model, performing simulation calculation and accumulation calculation according to a Monte Carlo simulation algorithm to obtain and store a first simulation daily load curve, and recording the first current simulation times;
judging whether the first current simulation times are not less than the set simulation times;
and when the first current simulation times are not less than the set simulation times, carrying out average calculation on all the first simulation daily load curves so as to obtain a predicted daily load curve.
5. A charging station load prediction device is characterized by comprising an acquisition statistic unit, a model fitting unit and a simulation calculation unit,
the acquisition and statistics unit is used for acquiring a charging data group of an electric vehicle in a charging station and a preset prediction condition, and performing data statistics on the charging data group according to a statistical requirement to acquire a first charging behavior data group; the prediction conditions comprise statistical requirements, model fitting parameters and simulation parameters;
the model fitting unit is used for respectively performing Gaussian fitting on each data group in the first charging behavior data group according to the model fitting parameters so as to obtain a charging behavior Gaussian model group through fitting; wherein each gaussian model in the charging behavior gaussian model group corresponds to each data group in the first charging behavior data group one to one: the method includes the steps of performing gaussian fitting on each data set in a first charging behavior data set according to the model fitting parameters, so as to obtain a charging behavior gaussian model set through fitting, and specifically includes: acquiring a Gaussian distribution probability density function model, a first fitting segment number, a minimum Gaussian fitting segment number, a maximum Gaussian fitting segment number, a fitting stopping precision threshold value and a fitting stopping precision threshold value from the model fitting parameters; the initial value of the first fitting segment number is the minimum Gaussian fitting segment number; taking the first fitting segment number as the segment number, performing Gaussian fitting on one data set in the first charging behavior data set according to a Gaussian distribution probability density function model by adopting a nonlinear least square method, so as to calculate and obtain a corresponding first charging Gaussian model, a first fitting precision and a first fitting precision improvement value, and taking the first charging Gaussian model as a first charging Gaussian distribution function; judging whether the first fitting precision is greater than the fitting stopping precision threshold value or not; if so, outputting the first charging Gaussian distribution function as a charging behavior Gaussian model; if not, judging whether the first fitting precision improvement value is smaller than the fitting stopping precision threshold value or not; if the first charging Gaussian distribution function is smaller than the first charging Gaussian distribution function, outputting the first charging Gaussian distribution function as a charging behavior Gaussian model; otherwise, taking the next value of the current first fitting segment number as a new first fitting segment number within the preset fitting segment number range; the fitting segment number range is from the minimum Gaussian fitting segment number to the maximum Gaussian fitting segment number; taking the new first fitting segment number as the segment number, and performing Gaussian fitting on the data set according to a Gaussian distribution probability density function model by adopting a nonlinear least square method, so as to calculate and obtain a corresponding second charging Gaussian model, a corresponding second fitting precision and a second fitting precision improvement value; judging whether the second fitting precision is greater than the fitting stopping precision threshold value or not; if so, outputting the second charging Gaussian distribution function as a charging behavior Gaussian model; if not, judging whether the second fitting precision improvement value is smaller than the fitting stopping precision threshold value or not; if the charging behavior is smaller than the first charging behavior, outputting the second charging Gaussian distribution function as a charging behavior Gaussian model;
and the simulation calculation unit is used for carrying out simulation according to the charging behavior Gaussian model and the simulation parameters through a Monte Carlo simulation algorithm so as to obtain a predicted daily load curve of the charging station load.
6. The charging station load prediction device according to claim 5, further comprising a data storage unit, wherein the data storage unit is configured to receive the charging data set of the electric vehicle in the charging station sent by the intelligent acquisition terminal and store the charging data set; the charging data set comprises a first charging data set for each charging pile to supply power each time and the number of vehicles entering the charging station in different time periods; the first charging data set comprises fast charging waiting time, slow charging waiting time, starting charging time, ending charging time, charging station entering time, charging station leaving time, battery capacity and initial state of charge.
7. The charging station load prediction device according to claim 5 or 6, wherein the obtaining statistics unit is further configured to remove invalid values in the charging data set.
8. A system for predicting a load of a charging station, the system comprising a prediction device of the load of the charging station and the charging station, the prediction device being communicatively connected to the charging station, the charging station comprising a plurality of parking spaces, a plurality of charging posts and a plurality of intelligent acquisition terminals, wherein each parking space is provided with one charging post and one or more intelligent acquisition terminals, the prediction device being configured to perform the method for predicting the load of the charging station according to any one of claims 1 to 4.
9. The system of claim 8, wherein the charging station loads are predicted to include fast charging and slow charging stations.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110976969.2A CN113821911B (en) | 2021-08-24 | 2021-08-24 | Charging station load prediction method, device and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110976969.2A CN113821911B (en) | 2021-08-24 | 2021-08-24 | Charging station load prediction method, device and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113821911A CN113821911A (en) | 2021-12-21 |
| CN113821911B true CN113821911B (en) | 2022-05-17 |
Family
ID=78913550
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110976969.2A Active CN113821911B (en) | 2021-08-24 | 2021-08-24 | Charging station load prediction method, device and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113821911B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117648520B (en) * | 2024-01-29 | 2024-05-07 | 北京理工大学 | New energy automobile charging load analysis-based carbon emission calculation method and system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108215872A (en) * | 2017-12-01 | 2018-06-29 | 国网北京市电力公司 | Charging method, device, storage medium and the processor of electric vehicle |
| CN111160639A (en) * | 2019-12-21 | 2020-05-15 | 杭州电子科技大学 | Electric vehicle charging load prediction method based on user travel time-space distribution characteristics |
-
2021
- 2021-08-24 CN CN202110976969.2A patent/CN113821911B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108215872A (en) * | 2017-12-01 | 2018-06-29 | 国网北京市电力公司 | Charging method, device, storage medium and the processor of electric vehicle |
| CN111160639A (en) * | 2019-12-21 | 2020-05-15 | 杭州电子科技大学 | Electric vehicle charging load prediction method based on user travel time-space distribution characteristics |
Non-Patent Citations (1)
| Title |
|---|
| 电动汽车充电行为对地区电网潮流的影响;刘博;《中国优秀硕士学位论文全文数据库》;20210215;第8-40页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113821911A (en) | 2021-12-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Quirós-Tortós et al. | Statistical representation of EV charging: Real data analysis and applications | |
| Moreira et al. | A stochastic model to simulate electric vehicles motion and quantify the energy required from the grid | |
| CN109103878B (en) | Electric automobile group ordered charging method and power utilization optimization method for power distribution network | |
| CN108376985B (en) | Method for establishing charging load model of electric automobile and calculating node voltage | |
| CN117040028B (en) | Control strategy optimization method and system for optical storage and charging micro-grid of industrial and commercial park | |
| Akil et al. | Impact of electric vehicle charging profiles in data-driven framework on distribution network | |
| CN113821911B (en) | Charging station load prediction method, device and system | |
| CN115158097B (en) | Charging pile management method and system based on Internet of things and storage medium | |
| CN115954910B (en) | Distributed energy storage control method and system based on energy optimization | |
| CN107340476A (en) | The electrical state monitoring system and electrical state monitoring method of battery | |
| Ni et al. | A methodology to model daily charging load in the EV charging stations based on Monte Carlo simulation | |
| CN111325402A (en) | Method for predicting charging behavior of electric vehicle user based on BP neural network | |
| CN115700717A (en) | Power distribution analysis method based on electric automobile power consumption demand | |
| CN104283288A (en) | Charging method and device of charging station | |
| Hasan et al. | Probabilistic Modelling of Electric Vehicle Charging Demand based on Charging Station Data | |
| CN113119796A (en) | Electric vehicle residual charging time prediction method and system based on cloud sparse charging data | |
| Xu et al. | A framework for assessing the impact of plug-in electric vehicle to distribution systems | |
| CN114580251A (en) | Method and device for analyzing charging load of electric vehicle in power distribution area | |
| Jiang et al. | Uncertainty Quantification Assessment for Power Systems with Fast Electric Vehicle Charging Stations in Dynamic Stability Analysis | |
| CN115856655A (en) | New energy automobile battery real-time state evaluation method, storage medium and electronic equipment | |
| CN115542235A (en) | Method, device and equipment for determining metering error of charging gun and storage medium | |
| CN110020505B (en) | Small data-based large-scale charging pile planning model modeling method | |
| Yusuf et al. | Analyses and Applications of Plug-in Electric Vehicle Charging Stations' User Behavior in a Large University Campus Community | |
| Botero et al. | Demand forecasting associated with electric vehicle penetration on distribution systems | |
| CN113125982B (en) | Battery life prediction method and device |
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 |