CN110570014B - Electric vehicle charging load prediction method based on Monte Carlo and deep learning - Google Patents
Electric vehicle charging load prediction method based on Monte Carlo and deep learning Download PDFInfo
- Publication number
- CN110570014B CN110570014B CN201910725799.3A CN201910725799A CN110570014B CN 110570014 B CN110570014 B CN 110570014B CN 201910725799 A CN201910725799 A CN 201910725799A CN 110570014 B CN110570014 B CN 110570014B
- Authority
- CN
- China
- Prior art keywords
- charging
- electric
- load
- time
- initial
- 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
- 238000007600 charging Methods 0.000 title claims abstract description 316
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000013135 deep learning Methods 0.000 title claims abstract description 26
- 238000000342 Monte Carlo simulation Methods 0.000 claims abstract description 24
- 238000004364 calculation method Methods 0.000 claims abstract description 14
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 10
- 230000029305 taxis Effects 0.000 claims abstract description 9
- 238000009826 distribution Methods 0.000 claims description 22
- 239000011159 matrix material Substances 0.000 claims description 16
- 230000006870 function Effects 0.000 claims description 15
- 238000005070 sampling Methods 0.000 claims description 13
- 238000004458 analytical method Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 239000000446 fuel Substances 0.000 claims description 4
- 230000007423 decrease Effects 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- QBPFLULOKWLNNW-UHFFFAOYSA-N chrysazin Chemical compound O=C1C2=CC=CC(O)=C2C(=O)C2=C1C=CC=C2O QBPFLULOKWLNNW-UHFFFAOYSA-N 0.000 claims 1
- 238000001914 filtration Methods 0.000 claims 1
- 238000012614 Monte-Carlo sampling Methods 0.000 abstract description 4
- 238000011161 development Methods 0.000 description 6
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 description 4
- 229910052744 lithium Inorganic materials 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 238000010280 constant potential charging Methods 0.000 description 1
- 238000010277 constant-current charging Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Engineering & Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Computational Linguistics (AREA)
- General Business, Economics & Management (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Marketing (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Primary Health Care (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
- Secondary Cells (AREA)
- Electric Propulsion And Braking For Vehicles (AREA)
Abstract
The invention provides an electric vehicle charging load prediction method based on Monte Carlo and deep learning. The prediction method comprises the following steps: firstly, dividing the electric automobile into 4 types of electric buses, electric taxis, electric private cars and electric official cars according to the characteristics of the electric automobile, establishing a probability model of load influence factors, and further obtaining calculation models of charging power of different types of electric automobiles; secondly, extracting the initial charge state, the initial charging time and the like of the electric automobile by adopting a Monte Carlo simulation method to calculate the charging load of the electric automobile at each moment according to the reserve prediction result of the electric automobile; and finally, according to the electric vehicle charging load at each moment obtained by Monte Carlo sampling, carrying out deep learning and prediction on the electric vehicle charging load by adopting an LSTM deep learning algorithm, thereby obtaining an electric vehicle charging load curve. The charging load prediction method has better scientificity and objectivity.
Description
Technical Field
The invention belongs to the field of power systems, and particularly relates to an electric vehicle charging load prediction method based on Monte Carlo and deep learning.
Background
Due to the increasingly prominent problems of energy safety and environmental pollution, new energy resources are vigorously developed in recent years, wherein electric automobiles are developed rapidly, traditional fuel automobiles are gradually replaced, the electric automobiles become main vehicles in the future, and the electric automobiles have wide development prospects. Compared with other low-carbon loads, the electric automobile has the basis of large-scale application and becomes an important component of the power grid load. When the electric automobile is applied in a large scale, the influence of the electric automobile on a power grid cannot be ignored. Therefore, the development rule and the charging rule of the electric automobile are fully considered, and the method has important significance for long-term development of the urban power distribution network. The reserve and the proportion of the electric vehicle are predicted, the charging load is modeled, and the future daily load prediction result is provided to be the basis for carrying out the influence analysis of the electric vehicle access on the power grid, the planning and the control operation of the power distribution network, the two-way interaction of the electric vehicle and the power grid, the coordination research of the electric vehicle and other energy sources, traffic and other systems.
At present, in the aspect of load prediction of electric vehicles, some indexes and methods have been proposed, for example, common load prediction methods include a unit consumption method, a trend analysis method, an elastic coefficient method, a regression analysis method, a time series method, a gray model method, a neural network method, a delphite method, an expert system method, and a preferred combination analysis method.
The existing electric automobile charging load prediction research still has the following defects: the universality is not strong. Most researches only consider simple factors of modeling conditions such as charging travel distance, charging starting time, starting charge state and the like, and a theoretical load model covering various key factors and various vehicle types is not available. And the traditional method and the shallow learning method have insufficient self-adaptive capacity, insufficient cognitive capacity on nonlinear characteristic loads and low prediction precision.
Therefore, the method for predicting the charging load of the electric vehicle still needs to be improved.
Disclosure of Invention
In order to solve the technical problem, the invention provides an electric vehicle charging load prediction method based on Monte Carlo and deep learning. According to the method, an influence probability model of load factors is established, charging load calculation models of different types of electric vehicles are established, charging loads at all times are obtained by adopting a Monte Carlo simulation method, and then a charging load prediction curve of the electric vehicle is obtained by adopting a deep learning algorithm.
The invention is realized by adopting the following technical scheme:
an electric vehicle charging load prediction method based on Monte Carlo and deep learning comprises the following steps:
firstly, electric automobiles are divided into 4 types of electric buses, electric taxis, electric private cars and electric business cars according to the characteristics of the electric automobiles, a probability model of load influence factors is established, and then calculation models of charging power of the electric automobiles of different types are obtained.
And secondly, extracting the initial charge state, the initial charge time and the like of the electric automobile by adopting a Monte Carlo simulation method according to the reserve prediction result of the electric automobile to calculate the charge load of the electric automobile, and obtaining the charge load at each moment.
And finally, predicting the charging load of the electric vehicle by adopting an LSTM deep learning algorithm according to the charging load of the electric vehicle at each moment obtained by Monte Carlo sampling, thereby obtaining a charging load curve of the electric vehicle.
In the above technical solution, further, the establishing a load influence factor probability model in step 1) to obtain a calculation model of the charging power of the electric vehicle includes:
(1) the charging load of the electric vehicle is influenced by various factors, and the main influencing factors are initial charging time, daily driving mileage, charging time, electric vehicle holding capacity, initial state of charge (SOC), battery capacity and the like.
The method comprises the following steps of dividing the electric automobile into 4 types of electric buses, electric taxis, electric private cars and electric business cars according to the characteristics of the electric automobile, constructing an initial charging time model, and obtaining the initial charging time of the electric automobile through data fitting treatment, wherein the initial charging time of the electric automobile meets the normal distribution shown in the following formula:
in the formula: t is initial charging time, namely the end time of the last trip; mu.saAnd σaThe expectation and the standard deviation of the initial charging time are respectively, and the expectation and the standard deviation of different types of electric vehicles are different because the normal distribution results of the initial charging time fitted by the four types of vehicles are different;
(2) a daily driving mileage model is constructed, the daily driving mileage is an important index of the driving characteristics of the electric automobile, the power consumption of the automobile in one day is reflected, and the charging time of the electric automobile is further influenced. Similarly, the trip characteristic of the traditional fuel vehicle is used for replacing the trip characteristic of the electric vehicle for analysis, the daily driving mileage of the electric vehicle obeys log-normal distribution, and the probability density function is as follows:
in the formula: s is the daily mileage in km; mu.sbAnd σbThe expectation and variance, respectively, of the logarithm lns of the driving range s, vary with the driving characteristics of different types of electric vehicles.
(3) And constructing an initial charge state model, wherein the power demand required by the electric automobile to charge is changed along with the time in a charging period. To determine the charging load of the electric vehicle, the state of charge at the battery charging start time, i.e., the starting SOC, must be obtained. The initial SOC is a random function of the driving distance of the electric automobile after the last charging, and the value range is 0-100%. Assuming that the state of charge of the electric vehicle linearly decreases with the driving range, the state of charge at the initial charging time can be estimated from the driving range of the vehicle, and is expressed by a probability density function:
SOC＝(SOC0-s/smax)×100％
in the formula: sOCRepresents the starting SOC of the battery charge; sOC0Representing the state of charge value of the battery after the last charge. S is more dispersed because the charging time and the charging place of the electric automobile are more dispersedOC0Often not 1; smaxRepresents the maximum number of miles that can be driven after the battery is fully charged, in km.
(4) A charging duration model is constructed, at present, most of electric automobiles use lithium batteries, and the charging process of the lithium batteries is a constant-voltage and constant-current two-stage charging process. Assuming that the whole charging process is constant power, the charging duration has the following two calculation methods:
calculating the charge duration based on the state of charge, the following equation applies:
in the formula: te is charging time length, and the unit is h; u is the battery capacity, and the unit is kW.h; p is charging power, and the unit is kW; eta is charging efficiency;
calculating the charging time based on the daily mileage, as shown in the following formula:
in the formula: w100The unit is (kW.h)/hundred kilometers for the power consumption of the automobile per 100 km.
Further, the step 2) of calculating the charging load at each time includes:
(1) dividing travel characteristics of different types of electric automobiles: the method comprises the following steps of dividing the electric automobiles into 4 types of buses, taxis, private cars and business cars according to different purposes, and specifically analyzing the travel characteristics of the electric automobiles of different types to obtain parameters of a load prediction model, wherein the method comprises the following steps:
electric bus
The running characteristics of the bus are relatively very fixed, and a shift operation system is mainly adopted. According to the data in the document electric vehicle charging load prediction based on Monte Carlo simulation, the daily driving mileage of the bus is about 70 km. Considering safe operation, the operation requirement of the electric bus is difficult to be met by one-day charging, and two-day charging is needed. The operation time and the route of the bus are relatively centralized, and centralized charging can be performed. The rapid charging is carried out in the noon time period, and the conventional charging is carried out after work in the evening. Generally, the charging period of a bus is 9.30 to 16.00, 23: 00 to the next day 05: 00, respectively obeying normal distribution, wherein specific distribution parameters can be obtained according to urban research data.
Electric taxi
The operating time of a taxi is approximately 06: 00- -24: 00. according to data in the document "special motor vehicle traffic trip characteristics and development strategies in Beijing City", the daily driving mileage of a taxi is about 400 km. Like an electric bus, an electric taxi generally adopts a mode of charging twice a day, and the charging time is selected to change shifts in the noon and at night. The electric taxi selects a quick charging mode because the rest time of the taxi is limited but the electric quantity needs to be supplemented in time. According to the above analysis, the charging period of the taxi is 02: 00-05: 00. 11: 30-14: 30, respectively, follow a normal distribution.
Electric private car
Compared with buses and taxis, the running characteristic of the private car is more random and arbitrary. The electric private car is charged once a day, and the charging period is divided into 09 in the morning: 00-12: 00. 14 in the afternoon: 00- -17: 00 and 19 in the evening: 00 to the next day 07: 00. charging in the parking lot of a work unit in the morning and afternoon, and charging in the parking lot of a residential area after work in the evening. A quick charging mode is selected for charging in a unit parking lot, and a conventional charging mode is selected for charging in a residential parking lot. According to the data in the literature, the daily driving mileage of a private car is 40km, regardless of long distance travel.
Electric commercial vehicle
The business car is mainly used for daily business trip, and if long-distance trip is not considered, the driving characteristics of the business car are similar to those of a private car. The charging time of a bus is generally that the bus is charged in a unit parking lot after work at night. The bus can be charged once a day, and a conventional charging mode is adopted, wherein the charging time period is 19: 00 to the next day 07: 00.
(2) predicting the holding capacity of various types of electric automobiles: the data in 2020 shown in Chinese automobile industry development report 2012 is used as a cardinal number, and the quantity of various types of automobiles is predicted according to the proportion of different types of electric automobiles,
calculating the charging load of the electric automobile based on Monte Carlo simulation:
making reasonable assumptions:
a. the initial charging time, the daily mileage and the charging power of each type of electric automobile are mutually independent random variables;
b. the charging power of each type of electric automobile is regarded as a constant power model;
c. all vehicles are fully charged each time;
d. and the last trip end time of all the vehicles is the initial charging time of the vehicle.
(3) And selecting a Monte Carlo method simulation method to sample different types of electric automobiles, and extracting initial charging time and daily driving mileage to obtain the charging time and charging power of a single electric automobile. Then accumulating the charging power of all the electric automobiles to obtain a charging load; the charging load curve can be obtained by using the time as an abscissa and the charging load at each time as an ordinate. The charging load at time j is then:
in the formula: ptotal，jThe charging load at time j is in kw; n is a radical ofb，Nt，Ns，NwThe total number of the electric bus, the taxi, the private car and the commercial car is respectively, and the unit is a vehicle; dividing a day into T time periods; pibj,Pibj,Pibj,PibjThe charging power of different types of electric automobiles at the moment j is in kW.
(4) Because the charging power is constant, the charging load is only related to the charging time, the charging time of the electric automobile is taken as a sampling constraint condition of the initial charging time, namely, the sampling range of the initial charging time can be selected, and then the initial charging time is extracted, so that the charging load at the corresponding moment of each sampling point is obtained; the calculation method of the charging time is as follows:
the electric automobile has two modes of conventional charging and quick charging, and the specific charging condition is relatively complex. For convenient modeling, assuming that the automobile preferentially selects conventional charging, calculating the charging time based on the daily mileage during conventional charging, extracting the daily mileage by using a Monte Carlo method, and calculating the required charging time; and selecting a sampling range of the initial charging time on the premise of meeting the constraint condition of the time length required by charging, then extracting the initial charging time, and carrying out load calculation. And when the quick charging mode is selected, the charging time length is calculated based on the charge state, the initial charging time is extracted based on the Monte Carlo method, the remaining charging time length in the charging time period and the charging time length meeting the driving requirement when the next charging is carried out are calculated, and the short time of the remaining charging time length and the charging time length is taken as the actual charging time length.
Further, step 3) of predicting the charging load of the electric vehicle by using an LSTM deep learning algorithm comprises the following steps:
(1) before the electric vehicle charging load prediction is carried out, normalization processing needs to be carried out on input data so as to ensure that network training and application can have good results. Normalized by min-max, the formula is as follows:
in the formula: xoRepresenting the electric vehicle load matrix after standardization; x represents an original electric vehicle load matrix, the matrix has only one row, and the matrix corresponds to the load value at each moment; x is the number ofmin、xmaxRespectively representing the minimum value and the maximum value of the load in the original electric vehicle load matrix.
(2) The storage of the historical charging load information is determined. I.e. the output of the forgetting gate:
ft＝sigmoid(Wf·[ht-1,xt]+bf)
in the formula: f. oftIndicating a forgotten gate output; wfA weight coefficient representing the weight from the input to the forgetting gate; h ist-1An output representing a previous time; x is the number oftRepresenting the current time input of the network; bfIndicating forgetting the gate bias.
(3) The storage of the new charging load information is determined. The first part is input gate output obtained through a sigmoid function; the other part is a new candidate vector established by tanh:
it＝sigmoid(Wi·[ht-1,xt]+bi)
in the formula: wiRepresenting an input gate weight parameter; biRepresenting input gate bias;
represents a candidate value of the charge load state of the cell, WCCalculating a weight parameter representing a candidate value of the cell charging load state; bCAnd calculating the bias of the candidate value representing the charging load state of the cell. The cell represents the cell at this time, which can be determined from the previous monte carlo sample point interval;
(4) updating the cell charge load state:
(5) and determining the charging load output at the current moment. In LSTM, the cell charge load state is processed by the tanh function and filtered by the output gate to obtain the final output:
ot＝sigmoid(Wo·[ht-1,xt]+bo)
ht＝ot*tanh(Ct)
in the formula: otRepresents the output of the output gate; woRepresenting a weight parameter input to the output gate; boRepresents the output gate offset; h istRepresenting the output at the current time.
(6) And (3) processing the charging load sample data of the electric vehicle obtained by Monte Carlo simulation by adopting the steps (1) to (5) to finally obtain the charging load prediction result of the electric vehicle of the LSTM network.
(7) After the charging load of the electric automobile is predicted, the root mean square error prediction accuracy evaluation index is adopted:
the invention has the beneficial effects that:
the invention establishes an electric vehicle charging load prediction method based on Monte Carlo and deep learning, and calculation models of different types of electric vehicle charging power and charging loads of electric vehicles at all times can be obtained according to the method. The method is used for predicting the charging load of the electric vehicle in the future day based on Monte Carlo simulation and a deep learning algorithm, so that the overall trend of the charging load of the electric vehicle in the future is obtained. Simulation results show that the method provided by the invention can accurately predict the charging load at each moment, and the scientificity and objectivity of the method for predicting the charging load are effectively proved.
Drawings
FIG. 1 illustrates charging modes and charging time distributions of different types of electric vehicles;
FIG. 2 is a graph of 10-day electric vehicle charging load data obtained by Monte Carlo sampling;
FIG. 3 is a diagram of the LSTM architecture;
FIG. 4 data from day 10 predicted after learning using the first 9 days of data;
FIG. 5 shows the number of hidden layer nodes as 400LSTM load prediction results and actual values;
fig. 6 is a flowchart of a charging load prediction method based on monte carlo and deep learning according to the present invention.
Detailed Description
The invention is further described with reference to the following figures and specific embodiments.
The invention discloses an electric vehicle charging load prediction method based on Monte Carlo and deep learning, which comprises the following steps:
s01, constructing an initial charging time model, and obtaining that the initial charging time of the electric automobile meets normal distribution shown in the following formula through data fitting processing:
in the formula: t is the initial charging time, namely the end time of the last trip; mu.saAnd σaThe expectation and variance of the initial charging time are different from one another, and the expectation and variance of different types of electric vehicles are different from one another.
And S02, constructing a daily driving mileage model, wherein the daily driving mileage is an important index of the driving characteristics of the electric automobile, and reflects the power consumption of the automobile in one day so as to influence the charging time of the electric automobile. Similarly, the trip characteristic of the traditional fuel vehicle is used for replacing the trip characteristic of the electric vehicle for analysis, the daily driving mileage of the electric vehicle obeys log-normal distribution, and the probability density function is as follows:
in the formula: s is the daily mileage in km; mu.sbAnd σbThe expectation and variance, respectively, of the logarithm lns of the driving range s, vary with the driving characteristics of different types of electric vehicles.
S03, constructing an initial charge state model, wherein in a charging period, the power demand required by charging of the electric automobile is changed along with time. To determine the charging load of the electric vehicle, the state of charge at the battery charging start time, i.e., the starting SOC, must be obtained. The initial SOC is a random function of the driving distance of the electric automobile after last charging, and the value range is 0-100%. Assuming that the state of charge of the electric vehicle linearly decreases with the driving range, the state of charge at the initial charging time can be estimated from the driving range of the vehicle, and is expressed by a probability density function:
SOC＝(SOC0-s/smax)×100％
in the formula: sOCRepresents the starting SOC of the battery charge; sOC0Represents the state of charge value of the battery after the last charge, and s is the daily mileage. S is more dispersed because the charging time and the charging place of the electric automobile are more dispersedOC0Often not 1; smaxRepresents the maximum number of miles that can be driven after the battery is fully charged, in km.
S04, constructing a charging duration model, wherein at present, most of electric automobiles use lithium batteries, and the charging process of the lithium batteries is a constant-voltage and constant-current two-stage charging process. The entire charging process is assumed to be constant power. The charging time is restricted by more factors, and the charging time is calculated based on the charge state, so that the following formula is provided:
in the formula: t iseIs the charging time length with the unit of h; u is the battery capacity, and the unit is kW.h; p is charging power, and the unit is kW; η is the charging efficiency.
S05, calculating the charging time length based on the daily mileage, and then, as shown in the following formula:
in the formula: w100The unit is (kW.h)/hundred kilometers for the power consumption of the automobile per 100 km.
S06, dividing travel characteristics of different types of electric automobiles: the electric automobiles are divided into 4 types of buses, taxis, private cars and business cars according to different purposes, and the travel characteristics of the electric automobiles of different types are specifically analyzed to obtain parameters of the load prediction model. The charging modes and charging time distribution of the 4 types of electric automobiles are shown in the attached figure 1.
S07, predicting the holding capacity of various types of electric automobiles: the data in 2020 indicated in the Chinese automobile industry development report (2012) is used as a cardinal number, and the quantity of various types of automobiles is predicted according to the proportion of different types of electric automobiles. The predicted retention amount of different types of electric vehicles is shown in table 1.
TABLE 1 holdover prediction for different types of electric vehicles
| Type of vehicle | Specific gravity (%) | Type of vehicle | Specific gravity (%) |
| Electric bus | 65.52 | Electric private car | 11.24 |
| Electric taxi | 15.52 | Electric commercial vehicle | 7.72 |
S08, calculating the charging load of the electric automobile based on Monte Carlo simulation:
making reasonable assumptions:
a. the initial charging time, daily mileage and charging power of each type of electric automobile are mutually independent random variables;
b. the charging power of each type of electric automobile is regarded as a constant power model, and the 2 stages of constant voltage and constant current charging do not exist;
c. all vehicles are fully charged each time;
d. and the last trip ending time of all the vehicles is the initial charging time of the vehicle.
And selecting a Monte Carlo method simulation method to sample different types of electric automobiles, and extracting initial charging time and daily driving mileage to obtain the charging time and charging power of a single electric automobile. And then accumulating the charging power of all the electric automobiles to obtain the charging load at each moment. The charging load curve can be obtained by using the time as an abscissa and the charging load at each time as an ordinate. The charging load at time j is then:
in the formula: wherein, Ptotal，jThe charging load at time j is in kw; n is a radical ofb，Nt，Ns，NwThe total number of the electric bus, the taxi, the private car and the commercial car is respectively, and the unit is a vehicle; dividing a day into T time periods; pibj,Pibj,Pibj,PibjThe charging power of different types of electric automobiles at the moment j is in kW.
S09, because the charging power is constant, the charging load is only related to the charging time, the charging time of the electric automobile is used as a sampling constraint condition of the initial charging time, a sampling range of the initial charging time can be selected, then the initial charging time is extracted, and therefore the charging load of each sampling point at the corresponding moment is obtained; the calculation method of the charging time is as follows:
supposing that the automobile preferentially selects conventional charging, calculating the charging time based on the daily mileage during the conventional charging, extracting the daily mileage by using a Monte Carlo method, and calculating the required charging time; and selecting a sampling range of the initial charging time on the premise of meeting the constraint condition of the time length required by charging, then extracting the initial charging time, and carrying out load calculation. And when the quick charging mode is selected, the charging time length is calculated based on the charge state, the initial charging time is extracted based on the Monte Carlo method, the remaining charging time length in the charging time period and the charging time length meeting the driving requirement when the next charging is carried out are calculated, and the short time of the remaining charging time length and the charging time length is taken as the actual charging time length. The charging load data of the electric vehicle obtained by the Monte Carlo simulation for 10 days is shown in the attached figure 2.
S10, before the electric vehicle charging load is predicted, normalization processing needs to be carried out on input data to ensure that network training and application can have good results. Normalized by min-max, the formula is as follows:
in the formula: xoRepresenting the electric vehicle load matrix after standardization; x represents an original electric vehicle load matrix, the matrix has only one row, and the matrix corresponds to the load value at each moment; x is the number ofmin、xmaxRespectively representing the minimum value and the maximum value of the load in the original electric vehicle load matrix.
S11. the structure diagram of the LSTM is shown in the attached figure 3. The storage of the historical charging load information is determined. I.e. the output of the forgetting gate:
ft＝sigmoid(Wf·[ht-1,xt]+bf)
in the formula: f. oftIndicating a forgotten gate output; wfA weight coefficient representing the weight from the input to the forgetting gate; x is the number oftRepresenting the current time input of the network; bfIndicating forgetting the gate bias.
And S12, determining storage of new charging load information. The first part is the input gate output obtained by a sigmoid function; the other part is a new candidate vector established by tanh:
it＝sigmoid(Wi·[ht-1,xt]+bi)
in the formula: wiRepresenting an input gate weight parameter; biRepresenting input gate bias;
representing a cell charge load state candidate; wCCalculating a weight parameter representing a candidate value of the cell charging load state; bCAnd calculating the bias of the candidate value representing the charging load state of the cell.
S13, updating the cell charging load state:
and S14, determining the charging load output at the current moment. In LSTM, the cell charge load state is processed by the tanh function and filtered by the output gate to obtain the final output:
ot＝sigmoid(Wo·[ht-1,xt]+bo)
ht＝ot*tanh(Ct)
in the formula: otRepresents the output of the output gate; woRepresenting a weight parameter input to the output gate; boRepresents the output gate offset; h istRepresenting the output at the current time.
And S15, after the steps are carried out on the electric vehicle charging load sample data obtained by Monte Carlo simulation, the electric vehicle charging load prediction result of the LSTM network is finally obtained.
S16, after the charging load of the electric automobile is predicted, the root mean square error prediction precision evaluation index is adopted:
application example:
there are 12.5 thousands of electric vehicles in a certain market, and the ratio of different types of electric vehicles is shown in table 1. Wherein the electric bus is charged twice a day, and the charging time period is respectively obeyed normal distribution N (13, 1)2) And N (23, 1)2) The power of the fast charge and the conventional charge is 108kW, 60kW, respectively. The daily mileage follows a lognormal distribution N (4.3, 0.34)2). The bus takes Biandy K9 as an example, the battery capacity of the electric bus is 324 kW.h, the endurance mileage is 250km, and the power consumption for driving 100km is 140 (kW.h)/hundred kilometers. The electric taxi can be charged twice a day, and the charging time intervals respectively follow normal distribution N (3.5, 1)2) And N (13, 1)2) The power for the rapid charging is 60 kW. The daily mileage follows a lognormal distribution N (5.2, 0.34)2). The model of Biandy E6 electric automobile is used as the analysis object of electric taxies, private cars and business cars. The capacity of the battery of the electric automobile is 82 kW.h, the endurance mileage is 400km, the power consumption for driving 100km is 20.5 (kW.h)/hundred kilometers, the charging power of the quick charging mode is 60kW, and the charging power of the conventional charging mode is 7 kW. The charging probabilities of the electric private car in the morning, the afternoon and the evening are respectively set to be 0.2,0.1 and 0.7, and the specific charging time respectively obeys normal distribution N (9.5 and 1)2)、N(14，12)、N(19，22) Daily mileage follows a lognormal distribution N (3.1, O.86)2). The charging period of the official vehicle follows normal distribution N (19, 2)2) Daily mileage follows a lognormal distribution N (3.1, O.86)2)。
Taking 15 minutes as a sampling point, adopting Monte Carlo simulation to extract 96 points in total to simulate the charging load of all electric vehicles in one day, and adopting load data of 9 days as sample data. The load data for day 10 predicted using the proposed LSTM deep learning algorithm is shown in fig. 4, compared to the day 10 sampled data obtained by monte carlo simulation is shown in fig. 5.
Therefore, the charging load of the electric vehicle obtained by LSTM deep learning prediction basically coincides with the charging load obtained by Monte Carlo sampling, and the error ratio is small. The LSTM prediction result error and the BP neural network prediction result error are shown in table 2, and it can be seen that the result obtained by deep learning is better than the shallow learning result of the BP neural network.
TABLE 2 comparison of LSTM and BP prediction errors
The peak value of the total charging load of the 4 types of electric automobiles is 14: 00-15: 00, about 3900 MW. This is because the electric vehicles are more selected to be charged in this period, and the rapid charging mode is often used for charging. The charging load curve shows that the fluctuation of the charging load in one day is severe, and the charging load has potential threat to the stability of the power grid. Therefore, the invention results in providing support for the power grid aspect to consider the corresponding guiding charging strategy to guide the charging behavior of the standard user.
Claims (3)
1. The method for predicting the charging load of the electric automobile based on Monte Carlo and deep learning is characterized by comprising the following steps of:
s1: dividing the electric automobiles into 4 types of electric buses, electric taxis, electric private cars and electric official cars according to the characteristics of the electric automobiles, establishing probability models of load influence factors, and further obtaining calculation models of charging power of different types of electric automobiles;
s2: according to the reserve prediction result of the electric automobile, extracting the initial charge state and the initial charge time of the electric automobile by adopting a Monte Carlo simulation method to calculate the charge load of the electric automobile at each moment;
s3: according to the electric vehicle charging load at each moment obtained by sampling by the Monte Carlo simulation method, deep learning and prediction are carried out on the electric vehicle charging load by adopting an LSTM deep learning algorithm, so that an electric vehicle charging load curve is obtained;
the step S1 specifically includes:
the method comprises the following steps of dividing the electric automobile into 4 types of electric buses, electric taxis, electric private cars and electric business cars according to the characteristics of the electric automobile, constructing an initial charging time model, and obtaining the initial charging time of the electric automobile through data fitting processing, wherein the initial charging time meets the following normal distribution:
in the formula, t is initial charging time, namely the end time of the last trip; mu.saFor the expectation of the initial charging time, σaIs the standard deviation of the initial charge time;
the daily mileage model is constructed, the trip characteristics of the traditional fuel vehicle are used for replacing the trip characteristics of the electric vehicle for analysis, the daily mileage of the electric vehicle obeys log-normal distribution, and the probability density function is as follows:
in the formula: s is the daily mileage in km; mu.sbExpectation of logarithm lns of daily driving range s, σbIs the standard deviation of the logarithm lns of the daily driving range s;
constructing an initial state of charge model, wherein in a charging period, the power demand required by the charging of the electric automobile changes along with time, and the charging load of the electric automobile is determined to obtain the state of charge at the battery charging starting moment, namely the initial SOC, wherein the initial SOC is a random function of the driving distance of the electric automobile after the last charging, and the value range is 0-100%; assuming that the state of charge of the electric vehicle linearly decreases with the driving range, the state of charge at the initial charging time is estimated from the driving range of the vehicle, and is expressed by a probability density function as follows:
SOC＝(SOC0-s/smax)×100％
in the formula: sOCRepresents the starting SOC of the battery charge; sOC0Represents the state of charge value of the battery after the last charge, s is the daily mileage, smaxIndicating feasibility after battery is fully chargedThe maximum mileage of driving, unit km;
a charging time period model is constructed, and if the whole charging process of the electric automobile is constant power, the charging time period has the following two calculation modes:
calculating the charge duration based on the state of charge, the following equation applies:
in the formula: t iseIs the charging time length with the unit of h; u is the battery capacity, and the unit is kW.h; p is charging power, and the unit is kW; eta is charging efficiency;
calculating the charging time based on the daily mileage, as shown in the following formula:
in the formula: w100The power consumption of the automobile is 100km per driving.
2. The method for predicting the charging load of the electric vehicle based on monte carlo and deep learning according to claim 1, wherein the method for calculating the charging load of the electric vehicle at each moment in step S2 comprises:
specifically analyzing travel characteristics of different types of electric automobiles to obtain parameters of a load prediction model;
predicting the holding capacity of various types of electric automobiles, predicting the quantity of various types of electric automobiles according to the proportion of different types of electric automobiles,
calculating the charging load of the electric automobile based on a Monte Carlo simulation method:
making reasonable assumptions:
a. the initial charging time, the daily mileage and the charging power of each type of electric automobile are mutually independent random variables;
b. the charging power of each type of electric automobile is regarded as a constant power model;
c. all vehicles are fully charged each time;
d. the last trip ending time of all the vehicles is the initial charging time of the vehicle;
selecting a Monte Carlo simulation method to sample different types of electric automobiles, obtaining the charging time and the charging power of a single electric automobile by extracting initial charging time and daily driving mileage, then accumulating the charging power of all the electric automobiles at each moment to obtain the charging load of the electric automobiles, wherein the charging load at the moment j is as follows:
in the formula: ptotal，jThe charging load at time j is in kw; n is a radical ofbIs the total number of electric buses, NtTotal number of electric taxis, NsTotal number of electric private cars, NwThe total number of the electric official vehicles is the unit of a vehicle; dividing a day into T time periods;
for the charging power of the electric bus at the moment j,
for the charging power of the electric taxi at the moment j,
charging power for the electric private car at time j,
the charging power of the electric official vehicle at the moment j is in kW;
because the charging power is constant, the charging load is only related to the charging time, the charging time of the electric automobile is taken as a sampling constraint condition of the initial charging time, namely, a sampling range of the initial charging time is selected, and then the initial charging time is extracted, so that the charging load at the corresponding moment of each sampling point is obtained; the calculation method of the charging time is as follows:
the electric automobile has two modes of conventional charging and quick charging, and if the automobile selects conventional charging, the charging time is calculated based on daily mileage during conventional charging; the method comprises the steps of selecting an electric automobile in a quick charging mode, calculating charging time based on a charge state, extracting initial charging time based on a Monte Carlo simulation method, calculating residual charging time in a charging time period and charging time meeting the driving requirement at the next charging, and taking the short time of the residual charging time and the charging time as the actual charging time.
3. The method for predicting the charging load of the electric vehicle based on monte carlo and deep learning according to claim 2, wherein the step S3 of predicting the charging load of the electric vehicle by using the LSTM deep learning algorithm comprises the following steps:
before the electric vehicle charging load prediction is carried out, normalization processing is carried out on input data, min-max standardization is adopted, and the formula is as follows:
in the formula: xoRepresenting the electric vehicle load matrix after standardization; x represents an original electric vehicle load matrix, the matrix has only one row, and the matrix corresponds to the load value at each moment; x is the number ofminRepresenting the minimum value x of the load in the original electric vehicle load matrixmaxRepresenting the maximum value of the load in the original electric vehicle load matrix;
determining the storage of the historical charging load information, namely the output of the forgetting gate:
ft＝sigmoid(Wf·[ht-1,xt]+bf)
in the formula: f. oftIndicating a forgotten gate output; wfA weight coefficient representing the weight from the input to the forgetting gate; h ist-1Represents the last oneOutputting the time; x is the number oftRepresenting the current time input of the network; bfIndicating a forgotten gate bias;
determining storage of new charging load information, wherein the first part is input gate output obtained through a sigmoid function; the other part is a new candidate vector established by tanh:
it＝sigmoid(Wi·[ht-1,xt]+bi)
in the formula: wiRepresenting an input gate weight parameter; biRepresenting input gate bias;
representing a cell charge load state candidate; wCCalculating a weight parameter representing a candidate value of the cell charging load state; bCRepresents the cell charging load state candidate calculation bias,
updating the cell charge load state:
determining the charging load output at the current moment, and in an LSTM deep learning algorithm, obtaining the final output after the cell charging load state is subjected to tanh function processing and output gate filtering:
ot＝sigmoid(Wo·[ht-1,xt]+bo)
ht＝ot*tanh(Ct)
in the formula: otRepresents the output of the output gate; woRepresenting a weight parameter input to the output gate; boRepresents the output gate offset; h istAn output representing a current time;
processing the charging load sample data of the electric vehicle obtained by the Monte Carlo simulation method by adopting the steps to finally obtain the charging load prediction result of the electric vehicle of the LSTM deep learning algorithm;
after the charging load of the electric automobile is predicted, the root mean square error prediction accuracy evaluation index is adopted:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910725799.3A CN110570014B (en) | 2019-08-07 | 2019-08-07 | Electric vehicle charging load prediction method based on Monte Carlo and deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910725799.3A CN110570014B (en) | 2019-08-07 | 2019-08-07 | Electric vehicle charging load prediction method based on Monte Carlo and deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110570014A CN110570014A (en) | 2019-12-13 |
| CN110570014B true CN110570014B (en) | 2022-04-01 |
Family
ID=68774762
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910725799.3A Active CN110570014B (en) | 2019-08-07 | 2019-08-07 | Electric vehicle charging load prediction method based on Monte Carlo and deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110570014B (en) |
Families Citing this family (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111291999A (en) * | 2020-02-21 | 2020-06-16 | 东南大学 | Dynamic wireless charging road section load modeling method based on hierarchical clustering |
| CN113554205A (en) * | 2020-04-26 | 2021-10-26 | 国网上海市电力公司 | Monte Carlo algorithm-based electric vehicle charging demand load research method |
| CN111626494A (en) * | 2020-05-22 | 2020-09-04 | 广东电网有限责任公司 | Power load prediction method, power load prediction device, computer equipment and storage medium |
| CN111626514B (en) * | 2020-05-29 | 2022-10-18 | 深圳供电局有限公司 | Electric vehicle charging load prediction method and device |
| CN111626527B (en) * | 2020-06-10 | 2023-02-03 | 太原理工大学 | Intelligent power grid deep learning scheduling method considering fast/slow charging/discharging form of schedulable electric vehicle |
| CN112001564B (en) * | 2020-09-04 | 2022-05-10 | 国网江西省电力有限公司电力科学研究院 | User leaving time prediction method based on long-short term memory network |
| CN112215415A (en) * | 2020-09-29 | 2021-01-12 | 长安大学 | Automobile charging load scene prediction method based on optimal quantile of probability model |
| CN112329984B (en) * | 2020-10-08 | 2023-02-07 | 国网河南省电力公司封丘县供电公司 | Electric vehicle optimal scheduling method based on electric vehicle cluster system division |
| CN112241596B (en) * | 2020-10-26 | 2022-09-02 | 国网上海市电力公司 | Urban power grid medium-long term production simulation method considering electric automobile |
| CN112330025B (en) * | 2020-11-06 | 2022-05-24 | 国网冀北电力有限公司张家口供电公司 | Prediction method of space-time charging load for urban electric vehicle |
| CN112613731B (en) * | 2020-12-21 | 2023-02-24 | 国网河南省电力公司经济技术研究院 | Energy efficiency evaluation method of electric vehicle charging station based on complex network theory |
| CN114742118B (en) * | 2020-12-23 | 2023-10-27 | 中国科学院广州能源研究所 | Electric automobile cluster charge-discharge load combination prediction method |
| CN112613682A (en) * | 2020-12-29 | 2021-04-06 | 国网江苏省电力有限公司宜兴市供电分公司 | Electric vehicle charging load prediction method |
| CN112785050A (en) * | 2021-01-12 | 2021-05-11 | 国网浙江省电力有限公司湖州供电公司 | Ordered charging scheduling method based on electric vehicle charging load prediction |
| CN113298298B (en) * | 2021-05-10 | 2023-12-29 | 国核电力规划设计研究院有限公司 | Short-term load prediction method and system for charging pile |
| CN113420953B (en) * | 2021-05-24 | 2022-08-02 | 国网上海市电力公司电力科学研究院 | Flexible load adjustability analysis method based on capacity credibility |
| CN113541168A (en) * | 2021-05-31 | 2021-10-22 | 国网江苏省电力有限公司电力科学研究院 | Electric vehicle cluster controllability determining method, scheduling method and system |
| CN113268815B (en) * | 2021-06-24 | 2023-04-07 | 南方电网科学研究院有限责任公司 | Power distribution network double-layer planning method, device, equipment and storage medium |
| CN113435663A (en) * | 2021-07-15 | 2021-09-24 | 国网冀北电力有限公司唐山供电公司 | CNN-LSTM combined load prediction method considering electric vehicle charging load influence |
| CN113595113A (en) * | 2021-07-23 | 2021-11-02 | 上海电机学院 | Power distribution network regional electric vehicle charging load prediction method and storage medium |
| CN113629709B (en) * | 2021-09-10 | 2024-05-17 | 广东电网有限责任公司 | Method, system, equipment and storage medium for acquiring charge and discharge model of electric vehicle |
| CN114239967A (en) * | 2021-12-20 | 2022-03-25 | 深圳供电局有限公司 | Electric vehicle load prediction method, system and storage medium |
| CN114498634B (en) * | 2022-02-17 | 2023-08-29 | 四川大学 | Electric automobile charging load prediction method based on ammeter data |
| CN114580789B (en) * | 2022-04-08 | 2024-05-03 | 东南大学 | Charging behavior-based multi-type electric vehicle load prediction method and system |
| CN116203448B (en) * | 2023-02-07 | 2023-10-03 | 江南大学 | Power battery residual life prediction method based on Monte Carlo and deep learning |
| CN116373666A (en) * | 2023-03-28 | 2023-07-04 | 无锡市政公用新能源科技有限公司 | Charging dispatching system for active power distribution of charging pile |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015022469A (en) * | 2013-07-18 | 2015-02-02 | 三菱重工業株式会社 | Power management device of electric automobile and power demand prediction method of electric automobile |
| CN107067110A (en) * | 2017-04-14 | 2017-08-18 | 天津大学 | Charging electric vehicle load spatio-temporal prediction method under car Road network pattern |
| CN107392400A (en) * | 2017-09-04 | 2017-11-24 | 重庆大学 | Meter and real-time traffic and the EV of temperature charging load spatial and temporal distributions Forecasting Methodology |
| CN109034498A (en) * | 2018-08-31 | 2018-12-18 | 国网上海市电力公司 | Consider the electric car charging load forecasting method of user's charge frequency and charge power variation |
-
2019
- 2019-08-07 CN CN201910725799.3A patent/CN110570014B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2015022469A (en) * | 2013-07-18 | 2015-02-02 | 三菱重工業株式会社 | Power management device of electric automobile and power demand prediction method of electric automobile |
| CN107067110A (en) * | 2017-04-14 | 2017-08-18 | 天津大学 | Charging electric vehicle load spatio-temporal prediction method under car Road network pattern |
| CN107392400A (en) * | 2017-09-04 | 2017-11-24 | 重庆大学 | Meter and real-time traffic and the EV of temperature charging load spatial and temporal distributions Forecasting Methodology |
| CN109034498A (en) * | 2018-08-31 | 2018-12-18 | 国网上海市电力公司 | Consider the electric car charging load forecasting method of user's charge frequency and charge power variation |
Non-Patent Citations (2)
| Title |
|---|
| 基于蒙特卡洛模拟的电动汽车充电负荷预测;乔文娟;《华电技术》;20180630;第40卷(第06期);第19-22页 * |
| 电动出租车充电负荷的时空随机分布特性;廖斌杰;《电力建设》;20170131;第38卷(第01期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110570014A (en) | 2019-12-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110570014B (en) | Electric vehicle charging load prediction method based on Monte Carlo and deep learning | |
| Li et al. | Prediction of electric bus energy consumption with stochastic speed profile generation modelling and data driven method based on real-world big data | |
| CN111400662B (en) | Space load prediction method considering charging requirements of electric automobile | |
| CN112557907A (en) | SOC estimation method of electric vehicle lithium ion battery based on GRU-RNN | |
| CN109934403A (en) | Charge load Analysis prediction technique in electric car resident region based on mathematical model | |
| CN109508826B (en) | Electric vehicle cluster schedulable capacity prediction method based on gradient lifting decision tree | |
| CN110232219B (en) | Electric vehicle schedulable capacity verification method based on data mining | |
| CN111047120B (en) | Electric automobile charging load prediction method under road-electricity coupling network | |
| Gao et al. | Charging load forecasting of electric vehicle based on Monte Carlo and deep learning | |
| CN103745110A (en) | Method of estimating operational driving range of all-electric buses | |
| CN109840640B (en) | Method and system for site selection of electric vehicle charging pile | |
| CN112001564B (en) | User leaving time prediction method based on long-short term memory network | |
| CN107067130B (en) | Rapid charging station capacity planning method based on electric vehicle Markov charging demand analysis model | |
| CN115860185A (en) | Power grid load prediction method considering charging modes of various types of electric automobiles in extremely high temperature weather | |
| CN115063184A (en) | Electric vehicle charging demand modeling method, system, medium, equipment and terminal | |
| CN112766677A (en) | Modeling method and charging system for residential electric vehicle charging demand based on fuzzy comprehensive evaluation method | |
| CN114282821A (en) | Scheduling method, system and equipment for sharing electric automobile | |
| Liu et al. | Load characteristics forecasting of Hubei power grid up to year 2030 with the development of electric vehicles | |
| CN111967686B (en) | Random load prediction method based on power utilization probability distribution function | |
| CN109919393A (en) | A kind of charging load forecasting method of electric taxi | |
| CN112765726A (en) | Service life prediction method and device | |
| CN110826813A (en) | Power grid optimization method based on charging difference requirements of household electric vehicle users | |
| He et al. | Multi-time simulation of electric taxicabs' charging demand based on residents' travel characteristics | |
| CN115470985A (en) | Method and system for predicting charging load of electric vehicle | |
| CN112365041A (en) | Urban electric vehicle holding capacity classification prediction method based on system dynamics |
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 |