CN110674988A - Urban charging station planning method based on electric vehicle travel big data - Google Patents
Urban charging station planning method based on electric vehicle travel big data Download PDFInfo
- Publication number
- CN110674988A CN110674988A CN201910900464.0A CN201910900464A CN110674988A CN 110674988 A CN110674988 A CN 110674988A CN 201910900464 A CN201910900464 A CN 201910900464A CN 110674988 A CN110674988 A CN 110674988A
- Authority
- CN
- China
- Prior art keywords
- charging
- station
- data
- state
- time
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000009826 distribution Methods 0.000 claims abstract description 46
- 230000007704 transition Effects 0.000 claims abstract description 28
- 238000004140 cleaning Methods 0.000 claims abstract description 16
- 230000002068 genetic effect Effects 0.000 claims abstract description 15
- 238000010276 construction Methods 0.000 claims abstract description 14
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 12
- 238000007418 data mining Methods 0.000 claims abstract description 7
- 238000013178 mathematical model Methods 0.000 claims abstract description 7
- 238000012216 screening Methods 0.000 claims abstract description 6
- 230000008569 process Effects 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 6
- 238000004088 simulation Methods 0.000 claims description 6
- 238000005065 mining Methods 0.000 claims description 5
- 239000011159 matrix material Substances 0.000 claims description 4
- 230000035772 mutation Effects 0.000 claims description 4
- 108090000623 proteins and genes Proteins 0.000 claims description 4
- 230000005611 electricity Effects 0.000 claims description 3
- 238000012423 maintenance Methods 0.000 claims description 3
- 230000001502 supplementing effect Effects 0.000 claims 1
- 230000008901 benefit Effects 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 8
- 239000013589 supplement Substances 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000010353 genetic engineering Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 239000000047 product Substances 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 208000000044 Amnesia Diseases 0.000 description 1
- 208000031091 Amnestic disease Diseases 0.000 description 1
- 206010011968 Decreased immune responsiveness Diseases 0.000 description 1
- 230000006986 amnesia Effects 0.000 description 1
- 210000000349 chromosome Anatomy 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000009827 uniform distribution Methods 0.000 description 1
Images
Classifications
-
- 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"
- G06Q10/047—Optimisation of routes or paths, e.g. travelling salesman problem
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
-
- 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
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Quality & Reliability (AREA)
- General Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Educational Administration (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Genetics & Genomics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Physiology (AREA)
- Primary Health Care (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
- Electric Propulsion And Braking For Vehicles (AREA)
Abstract
The invention relates to a method for planning an urban charging station based on electric vehicle travel big data, which comprises the following steps: 1) carrying out data cleaning and data mining on the big trip data of the electric vehicle to obtain an initial state rule, a state transition rule and a charging information rule; 2) simulating the initial state of the vehicle by adopting a roulette method, recursing the travel state at each moment by adopting a Markov principle, and obtaining the space-time distribution of the charging demand points by screening; 3) selecting candidate station sites according to the space-time distribution condition of the charging demand, integrating geographic factors and urban planning requirements, and calculating the selectable grade of each candidate station meeting the charging demand and not meeting the rate constraint; 4) establishing a mathematical model; 5) and solving by adopting a genetic algorithm, and carrying out genetic operation aiming at the candidate station addresses to obtain the most suitable construction grade configuration of the charging station. Compared with the prior art, the method has the advantages of improving the accuracy of obtaining the initial state rule, the state transition rule and the charging information rule, specifying the planning scheme and the like.
Description
Technical Field
The invention relates to a planning method for urban charging stations, in particular to a planning method for urban charging stations based on big trip data of electric vehicles.
Background
Under the background that fossil energy is gradually exhausted and the requirement for environmental protection is continuously improved, electric automobiles are increasingly popularized and applied due to the obvious advantages of high efficiency, energy conservation, zero emission and no pollution, and are developed rapidly. Perfecting the charging facility is one of the important foundations for electric automobile popularization. The reasonable planning of the position and the capacity of the charging station can meet the space-time requirements of charging of the electric automobile and obtain certain economic benefits.
The existing domestic literature on charging facility planning mostly uses the statistical data of the family travel questionnaire published by the U.S. department of transportation in the whole united states for demand prediction. In actual engineering construction, the charging station site selection mostly adopts questionnaire survey or traffic shutoff volume to carry out demand prediction. Because the time-space distribution of the charging requirements of the electric automobile is closely related to factors such as living habits of users, regional and human environments, economic and social development levels and the like, the modeling of the charging requirements of the electric automobile driven by the model is very difficult, and massive traveling and electricity utilization big data provide a new way for predicting the charging requirements of the electric automobile.
In the existing research, the characteristics of the electric automobile industry are also revealed by performing data analysis processing on operation data and operation data of the electric automobile industry by using a big data technology. However, there is a few research on applying the big data technology to the electric vehicle charging demand prediction to realize charging station planning. Patent CN 110189025a discloses a method for planning charging demands under different load growth modes, but this method cannot perform effective data screening when processing raw data, cannot perform level configuration of charging stations, and is not specific enough in planning results.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provide a method for planning an urban charging station based on big trip data of an electric vehicle.
The purpose of the invention can be realized by the following technical scheme:
a method for planning urban charging stations based on big trip data of electric vehicles comprises the following steps:
s1: data mining:
the big trip data of the electric vehicles are acquired by a new energy vehicle public data acquisition and monitoring research center, nearly 20 thousands of electric vehicles are accessed by the new energy vehicle public data acquisition and monitoring research center, and vehicle information including driving data, battery data, driving motor data, insulation resistance data, direction information, fuel consumption and the like is uninterruptedly uploaded at the frequency of 10-20 seconds/time in 24 hours. The initial sample data obtained by the big data platform, namely the trip big data, has large scale, multiple types, quick change and low value density, and needs to be cleaned and mined as follows.
The trip big data of the electric vehicle are cleaned, the digital characteristics of the initial state, the state transition and the charging information are mined, and the trip rules, namely the initial state rule, the state transition rule and the charging information rule, are summarized.
The main contents of data cleaning are as follows:
1.1) cleaning original data from a new energy automobile public data acquisition and monitoring research center, and extracting electric automobile charging and traveling information: the distance traveled, the remaining electric quantity (SOC), the position information (GPS, i.e., latitude and longitude), the vehicle state (whether traveling, parking, or charging), the charging power (UI), and the like of each vehicle, and the time corresponding thereto.
1.2) data interval adjustment. The original data density is too large, the time interval of the single-vehicle information is too small, the vehicle state (such as mileage, residual electric quantity and the like) of the electric vehicle is not obviously changed within 10-20 seconds according to the common sense, and the rule is not easy to see. And re-extracting the data for 15 minutes/time so as to analyze the travel rule of the electric automobile in the following process.
1.3) abnormal data elimination and missing data supplement. Eliminating inconsistent data and mutation data of position information or electric quantity information; and (5) filling up missing data by methods such as an interpolation method and the like.
1.4) sample availability judgment. If the data samples obtained through the processing satisfy the following conditions: and if the number of vehicles contained in the sample is not less than 10% of the actual reserved quantity, and the time span of the information of each vehicle is not less than 6 hours, the sample data can be used, otherwise, the sample data is cleaned again.
After cleaning, data mining is carried out, and the specific content is as follows:
and (3) mining the driving characteristics of the electric vehicle for the processed and judged data, and counting and summarizing the digital characteristics of the travel laws in the three aspects: initial state, state transition, charging information. Wherein the start state numerical characteristics include: probability distribution of travel time, probability distribution of day starting electric quantity, probability distribution of day starting position area and probability distribution of daily travel mileage; the state transition number features include: state transition probability distribution at each moment of working day and holiday; the charging information digital characteristics include: charging time probability distribution and charging position area probability distribution. And acquiring an initial state rule, a state transition rule and a charging information rule according to the digital characteristics.
S2: and (3) predicting the distribution of the charging demand:
simulating the starting time, the starting place, the electric quantity and the daily driving mileage of the vehicle by using a roulette method based on the starting state rule; then, from the initial state, the trip state at each moment is recurred by adopting a Markov principle based on a state transition rule, and the accumulated travel mileage reaches the daily travel mileage; and screening the charging demand points to obtain the space-time distribution of the charging demand points.
The principle and the using method of the round-robin bet and the Markov chain are as follows:
2.1) simulating the initial state of the vehicle by adopting a roulette method:
the roulette selecting method, also called proportion selecting operator, used by the invention has the basic idea that: the probability of each individual being selected is proportional to the magnitude of its fitness function value. If the population size is N and the fitness of the individual xi is f (xi), the selection probability of the individual xi is:
in the travel state simulation process, the probability is taken as the fitness, the sum of the probabilities of all states of the electric automobile at a certain time is 1, and any state randomly appears according to the probability, namely:
if the position status of the electric vehicle conforms to the principle of roulette, vehicle start status data at that time can be used.
Suppose that there are 3 positional states of the electric vehicle: h (located in residential), O (located in commercial), W (located in work), one of which is present for each vehicle at a time. Let the state probability of a certain vehicle on a certain working day at time t be p (h) 0.2, p (o) 0.5, and p (w) 0.3. Roulette selection can be simulated by:
① generates a uniformly distributed random number r within 0, 1.
②, when r < > is 0.2, the state at time t is H.
③, the state at time t is O if 0.2< r < > 0.7, and W if 0.7< r < > 1.
According to the initial state distribution probability obtained in the step S1, the travel initial time, the initial electric quantity, the initial area and the daily mileage of each car are obtained through simulation by the roulette method. After the initial area is selected, a place is randomly selected on the area according to uniform distribution to be set as an initial position.
2.2) adopting a Markov chain to recur the state at each moment:
markov, also known as anergy or amnesia. If x (t) is a discrete random variable, the Markov satisfies the equation:
P{x(t)＝x|x(tn)＝xn,...,x(t1)＝x1}＝P{x(t)＝x|x(tn)＝xn}
suppose a certain electric vehicle 4 on a certain working day: 00 to 8: the 00 states are H, H, H, H, W, respectively, then:
p { x (8) ═ W | x (7) ═ H } is the transition probability P for the state from H to W at the time of day t 7H-W 7Therefore, the equation is established, and the position state change of the electric automobile conforms to the Markov chain rule.
Not all electric vehicles in the area to be researched are connected to the platform, and the data need to be supplemented through simulation. The invention mainly aims at the daily travel of the electric passenger vehicles and the electric passenger cars in the common urban area to simulate, and does not consider special factors such as road repair, traffic jam and the like.
Suppose EK tThe state of the electric vehicle at time t is in the K (H, O, W inclusive) region, P (E)K t) Indicating an electric vehicle state as EK tProbability of (P)h-K tIndicating a certain motorThe probability that the vehicle is in the h (H, O, W inclusive) region at time t is changed to the K region at time t + 1. MtIs the state transition probability matrix at time t, MtIs a time-varying matrix, whose expression is:
the state probability at the next time is:
starting from the initial time and position obtained by the simulation in step S2, the state probability at the next time is obtained through the above formula recursion calculation. Then, the next time state is simulated by a roulette method, and the state E conforming to the electric automobile is found outK tRandomly selecting an area as the destination of the electric vehicle, and then selecting any point in road nodes of the area as the position P of the electric vehicle at the moment tt. Finally, searching P on the road network by utilizing Dijkstra methodt-1To PtAnd calculating the remaining capacity Qt. And when the accumulated mileage reaches the daily mileage simulated in 2.1) of the step S2, the trip is terminated.
If the vehicle needs to be charged at a certain place at a certain time, a collection of the time with the charging demand and the corresponding geographic position is called a demand point and is represented as [ t; (x, y) ], where t is time, x represents longitude and y represents latitude. A minimum charging threshold value is set, such as 20%, and charging is needed when the remaining electric quantity of the electric automobile is less than 20%. And screening out the moment when the residual electric quantity of each vehicle is lower than a threshold value, and recording the corresponding geographical position of each vehicle, thereby obtaining the space-time distribution of the demand points. And rearranging the screened demand point sets according to the moments, and drawing the spatial distribution of the demand points at each moment.
S3: selecting candidate station addresses:
on the basis of the known space-time distribution condition of the charging demand, selecting candidate station sites meeting the unreachable rate of the charging demand by comprehensively considering geographic factors and urban planning requirements, and calculating the optional grade of each candidate station meeting the rate constraint that the charging demand does not meet the rate constraint. The concrete contents are as follows:
selecting station sites meeting the urban planning requirements at dense demand points; and then calculating the rate eta of unsatisfied charging demand according to the following formula:
in which is composed of []Determining the number of elements, X, contained in the seti thA set of demand points, N, with the closest charging distance from the time t to the charging station ii thA set of points at which there is a charging demand at time t and charging station i can be reached.
Demand point set X with the shortest charging distance from time t to station ii th：
In the formula, GthIncluding all the charging demand points in the district at the time of the date h and the time t.
The point set which has a charging demand at the time t and can reach the station i is as follows:
selecting a station address satisfying the following constraint as a candidate station:
in the formula, REThe maximum kilometers of the electric automobile which can run after the full electric quantity is exhausted; the charging demand point n corresponds to the remaining electric quantity of the vehicle at the moment being QnDistance of shortest path to station YniDetermined by Dijkstra method (i station does not put Y on)ni＝∞)。
The selectable grade ranges for these candidate stations are calculated according to the following formula, i.e. the grade scale that a station may be built must meet the constraint of the rate at which the charging demand does not meet.
γ is the ratio of vehicles that have arrived at the station but have not been charged by the station. Xi(Si) The number of chargers in the station.
The present invention divides the level of the construction scale of the charging station into five levels, i.e., S, according to the above-mentioned constraintsiThe element belongs to {0,1,2,3,4}, and the 0 level is not established; level 1 is 8 chargers with 50-150 kw capacity; 2, 15 chargers with 150-250 kw capacity are arranged; the 3-level is 30 chargers with the capacity of 250-500 kw; the 4-level is 45 chargers with 500-750 kw capacity.
S4: establishing a model:
and establishing a mathematical model by taking the lowest comprehensive cost as a target function and taking the unreachable rate of the charging requirement, the unsatisfied rate of the charging requirement and cost constraint as constraint conditions.
4.1) objective function:
in the formula, FcostFor comprehensive cost, Ci vIs the capital that is consumed on average each year to build a charging station i:
in the formula, I belongs to I as the number of the candidate station, and I is the set of the candidate stations;
is the total floor area of the i-station building,
is the land price; class S of i station construction scaleiThe element belongs to {0,1,2,3,4} (hereinafter referred to as grade, 0 grade is not built; 1 grade is 8 chargers with 50-150 kw of capacity; 2 grade is 15 chargers with 150-250 kw of capacity; 3 grade is 30 chargers with 250-500 kw of capacity; 4 grade is 45 chargers with 500-750 kw of capacity); zi C(Si) Is i station capacity, CCThe cost per unit volume is the cost. m is the number of years for which the station i is expected to operate; r is0Is the return on investment.
4.2) constraint conditions:
a) geographical location unreachable constraint: for the i candidate station, in the charging service range (assuming that a certain vehicle is selected as soon as charging is needed, if the vehicle is closest to the i charging station, the vehicle is said to be in the service range of the i charging station), the ratio of the vehicles which have charging requirements and cannot reach the i station to be charged and have exhausted electric quantity is calculated.
Ratio of vehicles that have arrived but not charged at the station location:
b) cost constraint: total investment not greater than maximum budget Cmax：
S5: and (3) optimizing and solving: and solving an optimal scheme by adopting a genetic algorithm, setting a fitness function of the genetic algorithm by combining a mathematical model, and carrying out genetic operation on the candidate station addresses to find out the most suitable construction level configuration of the charging station.
The sequentially created states of n candidate stations (i.e., n Si) are used as chromosomes (genes) in a genetic algorithm, the plan is used as an individual, and a certain number of different individuals (plans) are used as a population.
The fitness function of the genetic algorithm is:
the invention eliminates individuals with high fitness (namely high cost), the individuals with low fitness preferentially survive, and retains own genes through heredity and variation, thereby finally generating the optimal solution. The optimal solution is the planning scheme with the lowest comprehensive cost and without exceeding the cost budget, and the optimal solution comprises the construction levels of the candidate charging stations.
The genetic algorithm is implemented as follows:
(1) and (3) encoding: for optimization variable SiBinary coding is used.
(2) Generating an initial population: at SiAnd randomly generating an initial population of a certain scale within a value range.
(3) Calculating a fitness value:
(4) genetic manipulation: the genetic manipulation strategy is two-by-two competitive selection, uniform crossing, uniform variation and optimal individual preservation.
(5) And (4) returning to the loop in step (3) until the iteration number is larger than the maximum algebra.
Compared with the prior art, the invention has the following advantages:
the application of big data in the field of electric vehicles is embodied, and the actual real-time travel data of the electric vehicles are used, so that a more accurate and reasonable planning scheme is obtained, and the problem that charging facilities are not matched with charging requirements is solved;
secondly, the method firstly carries out data cleaning and data mining on the trip big data, removes inconsistent data and mutation data of position information or electric quantity information, and supplements missing data, judges whether to carry out cleaning again by judging whether a sample can be used or not, and can further improve the accuracy of obtaining an initial state rule, a state transition rule and a charging information rule;
in the process of selecting the candidate station addresses meeting the unreachable rate of the charging demand, the selectable grade of each candidate station meeting the rate constraint that the charging demand does not meet is calculated, the number of charging stations can be more specifically planned, the optimal solution of a mathematical model is solved, and a planning scheme with the lowest comprehensive cost and without exceeding the cost budget can be obtained, wherein the scheme comprises the construction grade of each candidate charging station, and is favorable for the concretization of subsequent charging station planning;
fourthly, a roulette simulation method based on the Markov chain vividly and specifically presents the distribution conditions of the electric vehicles in the urban area in various types of areas on different dates and different moments, and then the space-time distribution of the charging demand is predicted;
and fifthly, the established minimum model of the total cost is simple and universal, is not limited by geographic environment factors, policy factors, population density, electric vehicle holding capacity and the like, and the genetic algorithm is simple and easy to solve.
Drawings
FIG. 1 is a schematic flow diagram of the process of the present invention;
FIG. 2 is a sectional view of an urban area according to an embodiment of the present invention;
fig. 3 is a digital characteristic result diagram of a driving rule of an electric vehicle in an embodiment of the present invention, where fig. 3(a) is a diagram of distribution probability of an electric vehicle travel time, an initial electric quantity, and a daily driving mileage, fig. 3(b) is a diagram of transition probabilities from working days, holidays to other states, and fig. 3(c) is a diagram of a probability distribution of a charging time;
FIG. 4 is a graph of a 12:00 distribution of demand points for a particular working day as simulated in an embodiment of the present invention;
FIG. 5 is a diagram illustrating candidate station numbers and distribution according to an embodiment of the present invention;
FIG. 6 is a flow chart of a genetic algorithm in an embodiment of the present invention;
fig. 7 is a graph of the fitness curve in an embodiment of the invention.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, shall fall within the scope of protection of the present invention.
Examples
In this embodiment, the method of the present invention is explained based on actual regional electric vehicle data and actual real-time electric vehicle trip data, and the specific data content is as follows:
taking a certain urban area with 37 square kilometers as an example, the population is 57 thousands, and the electric automobiles are 1.2 thousands (for simplification of calculation, the model is commonly set as the model of the E6 common in the market, the speed vEV is 40km/h, and the rated electric quantity WEV is 60 kW.h). The urban area is divided into 10 areas by taking main roads as boundaries, wherein the number of business areas is three (O1, O2 and O3), the number of residential areas is 4 (H1, H2, H3 and H4), and the number of working areas is 3 (W1, W2 and W3), as shown in FIG. 2.
As shown in fig. 1, the invention relates to a method for planning an urban charging station based on big trip data of an electric vehicle, which specifically comprises the following steps:
the method comprises the steps of firstly, cleaning big traveling data of the electric automobile, mining digital characteristics of an initial state, state transition and charging information, and summarizing a traveling rule. The specific content of the data cleaning process is as follows:
1) cleaning original data from a new energy automobile public data acquisition and monitoring research center, and extracting electric automobile charging and traveling information: the distance traveled, the remaining electric quantity (SOC), the position information (GPS, i.e., latitude and longitude), the vehicle state (whether traveling, parking, or charging), the charging power (UI), and the like of each vehicle, and the time corresponding thereto.
2) And adjusting the data time interval. The original data density is too large, the time interval of the single-vehicle information is too small, the vehicle state (such as mileage, residual electric quantity and the like) of the electric vehicle is not obviously changed within 10-20 seconds according to the common sense, and the rule is not easy to see. And re-extracting the data for 15 minutes/time so as to analyze the travel rule of the electric automobile in the following process.
3) And abnormal data elimination and missing data supplement. Eliminating inconsistent data and mutation data of position information or electric quantity information; and (5) filling up missing data by methods such as an interpolation method and the like.
4) And judging the availability of the sample. If the data samples obtained through the processing satisfy the following conditions: and if the number of vehicles contained in the sample is not less than 10% of the actual reserved quantity, and the time span of the information of each vehicle is not less than 6 hours, the sample data can be used, otherwise, the sample data is cleaned again.
After cleaning, data mining is carried out, and the specific content is as follows:
and (3) mining the driving characteristics of the electric vehicle for the processed and judged data, and counting and summarizing the digital characteristics of the travel laws in the three aspects: initial state, state transition, charging information. Wherein the start state numerical characteristics include: probability distribution of travel time, probability distribution of day starting electric quantity, probability distribution of day starting position area and probability distribution of daily travel mileage; the state transition number features include: state transition probability distribution at each moment of working day and holiday; the charging information digital characteristics include: charging time probability distribution and charging position area probability distribution.
The results of data cleaning and mining are shown in fig. 3. As can be seen from the initial state diagram, the travel characteristics of the electric passenger vehicle on working days and holidays are different, the electric passenger vehicle and the electric passenger car also have different travel characteristics on the same date type, and the travel characteristics of the electric passenger car are basically the same on the working days and holidays. The initial electric quantity distribution of the passenger car is approximate to normal distribution, the probability of 60% -80% is maximum, the proportion of the larger or smaller electric quantity is smaller, and the electric quantity distribution is approximate to symmetry. This is related to the randomness of the charging behavior of private users. The passenger car is managed in a unified mode, sufficient electric quantity is generally required to be guaranteed, SOC is at least more than 60%, and therefore the passenger car is mostly on a large electric quantity trip. The daily driving mileage of a private car is mostly within 30 kilometers, and long-distance travel can be increased in holidays; the daily driving mileage of the electric passenger car is mostly within 90 kilometers, the distance travel is more than that of a passenger car, but the driving mileage difference of working days and holidays is not large.
As can be seen from the state transition diagram, the transition probability from the same initial state to different states is different and changes along with the time; the transition probability from different initial states to the same state is different and changes along with time; the transition probability of the same initial state has different trend in different date types.
Step two, simulating the initial state of the vehicle by a roulette method according to the travel rule obtained in the step one; then, the travel state at each moment is recurred by adopting a Markov chain principle based on a state transition rule; finally, the charging demand points are screened to obtain the temporal distribution, and fig. 4 shows the distribution of the charging demand points at 12:00 of a certain working day.
Thirdly, selecting a station site meeting the urban planning requirement in an area with dense charging demand points; and selecting the station address meeting the charging demand unreachable rate constraint as a candidate station, and calculating the selectable grade meeting the charging demand unreachable rate.
The candidate station selection method comprises the following steps:
if the vehicle needs to be charged at a certain place at a certain time, a collection of the time with the charging demand and the corresponding geographic position is called a demand point and is represented as [ t; (x, y) ], where t is time, x represents longitude and y represents latitude. A minimum charging threshold value is set, such as 20%, and charging is needed when the remaining electric quantity of the electric automobile is less than 20%. And screening out the moment when the residual electric quantity of each vehicle is lower than a threshold value, and recording the corresponding geographical position of each vehicle, thereby obtaining the space-time distribution of the demand points. And rearranging the screened demand point sets according to the moments, and drawing the spatial distribution of the demand points at each moment.
Selecting station sites meeting the urban planning requirements at dense demand points; and then calculating the rate eta of unsatisfied charging demand according to the following formula:
in which is composed of []Determining the number of elements, X, contained in the seti thA set of demand points, N, with the closest charging distance from the time t to the charging station ii thA set of points at which there is a charging demand at time t and charging station i can be reached.
Selecting a station address satisfying the following constraint as a candidate station:
The selectable grade ranges for these candidate stations are calculated according to the following formula, i.e. the grade scale that a station may be built must meet the constraint of the rate at which the charging demand does not meet.
In the formula, grade S of i station construction scaleiE {0,1,2,3,4} (hereinafter abbreviated as level, level 0 is not established; level 1 is8 chargers with 50-150 kw capacity; 2, 15 chargers with 150-250 kw capacity are arranged; the 3-level is 30 chargers with the capacity of 250-500 kw; 4-level is 45 chargers with 500-750 kw capacity); γ is the ratio of vehicles that have arrived at the station but have not been charged by the station. That is, for the i-candidate station, within the charging service range (assuming that a charging station is selected nearby when a certain vehicle needs to be charged, if the vehicle is closest to the i-charging station, the vehicle is said to be within the service range of the i-charging station), the rate of the vehicle that has a charging demand but cannot reach the i-station for charging and is depleted in electric quantity is occupied. In the presence of known Ni thAccording to the constraint inequality of the rate of unsatisfied charging demand, the selectable grade of the rate of unsatisfied charging demand of the i station can be obtained.
The candidate station distribution selected by the present embodiment is shown in fig. 5.
And step four, establishing a mathematical model by taking the lowest comprehensive cost as a target and taking the unreachable rate of the charging requirement and the unsatisfied rate of the charging requirement as constraints. The relevant substation parameters are shown in table 1. The construction costs include fixed costs such as equipment and construction in addition to land costs.
Table 1 charging station basic parameters
And step five, solving the model by using a genetic algorithm, wherein a flow chart is shown in figure 6, and a fitness curve is shown in figure 7. The optimization result is shown in table 2, and the optimal level configuration of each candidate station obtained by planning is given. The table shows the optimal level combination of the 20 candidate charging stations and the geographical location coordinates of each charging station to be commissioned.
Table 2 preferred results
The final preferred result is that 17 stations from the 20 candidate stations can be built, the combination of the levels of the stations enables the lowest comprehensive cost, the total cost is minus 1549 ten thousand yuan, and the planning scheme can be profitable.
While the invention has been described with reference to specific embodiments, the invention is not limited thereto, and those skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope of the invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A method for planning urban charging stations based on big trip data of electric vehicles is characterized by comprising the following steps:
1) the method comprises the steps of obtaining big trip data of the electric automobile, carrying out data cleaning and data mining on the big trip data of the electric automobile, and obtaining an initial state rule, a state transition rule and a charging information rule;
2) based on an initial state rule, simulating an initial state of the vehicle by adopting a roulette wheel method, and based on a state transition rule, recurrently deducing a trip state at each moment by adopting a Markov principle from the initial state, accumulating the running mileage until the running mileage reaches the daily running mileage, and obtaining the space-time distribution of the running mileage by screening charging demand points;
3) according to the space-time distribution condition of the charging demand, combining geographic factors and urban planning requirements, selecting candidate station sites which meet the unreachable rate of the charging demand, and calculating the selectable grade of each candidate station which meets the rate constraint that the charging demand does not meet the rate constraint;
4) establishing a mathematical model by taking the lowest comprehensive cost as a target function and taking the unreachable rate of the charging requirement, the unsatisfied rate of the charging requirement and cost constraint as constraint conditions;
5) and solving the established mathematical model by adopting a genetic algorithm, and carrying out genetic operation aiming at the candidate station sites to obtain a charging station construction level planning scheme with the lowest comprehensive cost and without exceeding the cost budget.
2. The urban charging station planning method based on big traveling data of electric vehicles according to claim 1, wherein in step 1), the specific steps of data cleaning comprise:
111) cleaning original data from a new energy automobile public data acquisition and monitoring research center, and extracting electric automobile charging and traveling information, including the driving mileage, the remaining electric quantity, the position information, the vehicle state, the charging power and the time corresponding to each information of each automobile;
112) re-extracting the data extracted in the step 111) for 15 minutes/time;
113) eliminating inconsistent data and mutation data of position information or electric quantity information, and supplementing missing data by adopting an interpolation method;
114) judging the data sample obtained through the processing, if the number of vehicles contained in the sample is not less than 10% of the actual reserved quantity and the time span of each vehicle information is not less than 6 hours, using the sample data, otherwise, cleaning the data again;
the specific content of data mining is as follows:
and (3) mining the driving characteristics of the electric vehicle according to the data after the data cleaning step processing and judgment, and counting and summarizing the digital characteristics of the travel laws in the three aspects: and acquiring an initial state rule, a state transition rule and a charging information rule according to the digital characteristics.
3. The urban charging station planning method based on big trip data of electric vehicles according to claim 2, wherein the initial state digital characteristics comprise trip time probability distribution, daily initial electric quantity probability distribution, daily initial position area probability distribution and daily trip mileage probability distribution; the state transition digital characteristics comprise state transition probability distribution at each moment of working days and holidays; the charging information digital characteristics comprise charging time probability distribution and charging position area probability distribution.
4. The urban charging station planning method based on big trip data of electric vehicles according to claim 1, wherein the specific contents of the selected candidate sites are as follows:
selecting a station address satisfying the following constraint as a candidate station:
5. The method for planning the charging stations in the urban area based on the big traveling data of the electric vehicles as claimed in claim 4, wherein the selectable grade range of the candidate stations, namely the grade scale of the possible construction of a certain station, is calculated according to the following formula:
gamma is the ratio of vehicles that have arrived but have not been charged at the station location, where Xi(Si) The number of chargers S in the stationiBuilding scale grades for the i-station;
the rating of the construction scale of the charging station is divided into five levels, i.e. S, according to the above constraintsiThe element belongs to {0,1,2,3,4}, and the 0 level is not established; level 1 is 8 chargers with 50-150 kw capacity; level 2 of 15 chargers, 150-25A capacity of 0 kw; the 3-level is 30 chargers with the capacity of 250-500 kw; the 4-level is 45 chargers with 500-750 kw capacity.
6. The method for planning the charging station in the urban area based on the big data of electric vehicle traveling according to claim 1, wherein in step 4), the objective function of the established model is as follows:
in the formula, FcostFor comprehensive cost, Ci VThe capital that is consumed each year on average to build a charging station i is expressed as:
wherein I belongs to I as the number of the candidate station, I is the set of the candidate stations,
is the total floor area of the i-station building,
to ground price, SiFor building scale of i-station, Zi C(Si) Is i station capacity, CCM is the number of years the station is expected to operate, r0Is the return rate of investment.
7. The urban charging station planning method based on big data of electric vehicle travel according to claim 6, characterized in that n candidate stations are put into operation sequentially, that is, n S candidate stationsiAs a gene in the genetic algorithm, a scheme is taken as an individual, a certain number of different individuals are taken as a population, and the fitness function of the genetic algorithm is as follows:
and (3) enabling the fitness to be high, namely eliminating individuals with high cost, preferentially living individuals with low fitness, reserving genes of the individuals through heredity and variation, and finally generating an optimal solution, namely a planning scheme with the lowest comprehensive cost and without exceeding the cost budget, wherein the scheme comprises the construction grade of each candidate charging station.
8. The method for planning the charging stations in the urban area based on the big data of electric vehicle travel according to claim 1, wherein in the step 2), the concrete contents of simulating the initial state of the vehicle by adopting a roulette method are as follows:
if the population size is N and the fitness of the individual xi is f (xi), the selection probability P (x) of the individual xii) Comprises the following steps:
in the travel state simulation process, the probability P (x) is selectedi) For fitness, the sum of the probabilities of all states of the electric automobile at a certain time is 1, and any state randomly appears according to the probability, namely:
and judging whether the position state of the electric automobile conforms to the principle of a roulette method, and if so, using the initial state data of the automobile at the moment.
9. The urban charging station planning method based on big data of electric vehicle travel according to claim 8, characterized in that the specific contents of the Markov chain for recursion of the state at each moment are as follows:
if x (t) is a discrete random variable, the Markov satisfies the equation:
P{x(t)＝x|x(tn)＝xn,...,x(t1)＝x1}＝P{x(t)＝x|x(tn)＝xn}
in the formula, P is the state probability of a certain position of the electric automobile at a certain moment;
the position state of the electric vehicle is assumed to include three types: h is in the state of being located in the residential area, O is in the state of being located in the business area, W is in the state of being located in the working area, and each vehicle is in one state at each moment; suppose EK tThe state of the electric vehicle at time t is in a K region including H, O, W, P (E)K t) The state of an electric vehicle is EK tProbability of (P)h-K tM is the probability that a certain electric vehicle is changed from the state of being in the h region including H, O, W at the time t to the state of being in the K region at the time t +1tIs the state transition probability matrix at time t, then MtIs a time-varying matrix, and the expression is:
the state probability at the next time is:
10. the method for planning the charging stations in the urban area based on the big data of electric vehicle travel according to claim 6, wherein the constraint conditions of the established model comprise:
a) geographical location unreachable constraint: for the i candidate station, the unreachable rate η is the ratio of vehicles which have charging requirements but can not reach the nearest charging station, and the expression is as follows:
in the formula, Xi thA set of demand points, N, with the closest charging distance from the time t to the charging station ii thA point set which has a charging demand at the moment t and can reach a charging station i;
the rate γ, i.e., the rate of vehicles that have come to a station but have not been charged by the gate, is not satisfied, and is expressed as follows:
in the formula [ ·]To determine the number of elements contained in a set, Xi(Si) The number of chargers in the station is set;
b) cost constraint: total investment not greater than maximum budget Cmax：
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910900464.0A CN110674988A (en) | 2019-09-23 | 2019-09-23 | Urban charging station planning method based on electric vehicle travel big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910900464.0A CN110674988A (en) | 2019-09-23 | 2019-09-23 | Urban charging station planning method based on electric vehicle travel big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110674988A true CN110674988A (en) | 2020-01-10 |
Family
ID=69077300
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910900464.0A Pending CN110674988A (en) | 2019-09-23 | 2019-09-23 | Urban charging station planning method based on electric vehicle travel big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110674988A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112069635A (en) * | 2020-09-01 | 2020-12-11 | 上海钧正网络科技有限公司 | Battery replacement cabinet deployment method, device, medium and electronic equipment |
| CN116523087A (en) * | 2022-01-21 | 2023-08-01 | 广州蔚景科技有限公司 | Multi-factor-integrated charging station site layout optimization method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120203726A1 (en) * | 2011-02-09 | 2012-08-09 | Diego Klabjan | Information systems and methods for deployment of charging infrastructure in support of electric vehicles |
| CN109711630A (en) * | 2018-12-28 | 2019-05-03 | 郑州大学 | A kind of electric car fast charge station addressing constant volume method based on trip probability matrix |
| CN110189025A (en) * | 2019-05-30 | 2019-08-30 | 国网上海市电力公司 | Consider the electric automobile charging station programme acquisition methods that different load increases |
-
2019
- 2019-09-23 CN CN201910900464.0A patent/CN110674988A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120203726A1 (en) * | 2011-02-09 | 2012-08-09 | Diego Klabjan | Information systems and methods for deployment of charging infrastructure in support of electric vehicles |
| CN109711630A (en) * | 2018-12-28 | 2019-05-03 | 郑州大学 | A kind of electric car fast charge station addressing constant volume method based on trip probability matrix |
| CN110189025A (en) * | 2019-05-30 | 2019-08-30 | 国网上海市电力公司 | Consider the electric automobile charging station programme acquisition methods that different load increases |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112069635A (en) * | 2020-09-01 | 2020-12-11 | 上海钧正网络科技有限公司 | Battery replacement cabinet deployment method, device, medium and electronic equipment |
| CN112069635B (en) * | 2020-09-01 | 2024-05-14 | 上海钧正网络科技有限公司 | Method and device for deploying battery changing cabinet, medium and electronic equipment |
| CN116523087A (en) * | 2022-01-21 | 2023-08-01 | 广州蔚景科技有限公司 | Multi-factor-integrated charging station site layout optimization method |
| CN116523087B (en) * | 2022-01-21 | 2024-04-02 | 广州蔚景科技有限公司 | Multi-factor-integrated charging station site layout optimization method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103241130B (en) | Energy management method and system for electric bus charging and swap station | |
| CN109711630A (en) | A kind of electric car fast charge station addressing constant volume method based on trip probability matrix | |
| CN110288212B (en) | Improved MOPSO-based electric taxi newly-built charging station site selection method | |
| CN110968837B (en) | Method for locating and sizing electric vehicle charging station | |
| CN110189025B (en) | Electric vehicle charging station planning scheme acquisition method considering different load increases | |
| Jia et al. | A novel approach for urban electric vehicle charging facility planning considering combination of slow and fast charging | |
| CN108932561B (en) | Electric vehicle charging path selection method considering nonlinear charging function | |
| CN103745110B (en) | Method of estimating operational driving range of all-electric buses | |
| CN108269008B (en) | Charging facility optimization planning method considering user satisfaction and distribution network reliability | |
| CN104751249A (en) | Space load prediction method | |
| CN110826841A (en) | Charging station planning method considering user charging experience and power distribution network operation risk | |
| CN111046576B (en) | Electric private car charging load prediction method considering double-network information | |
| Zeng et al. | Optimization of future charging infrastructure for commercial electric vehicles using a multi-objective genetic algorithm and real travel data | |
| CN117035185A (en) | Electric vehicle charging station layout optimization method and system based on dynamic charging demand | |
| CN105262167A (en) | Intra-regional electric vehicle ordered charging control method | |
| CN110674988A (en) | Urban charging station planning method based on electric vehicle travel big data | |
| CN114048920A (en) | Site selection layout method, device, equipment and storage medium for charging facility construction | |
| CN115344653A (en) | Electric vehicle charging station site selection method based on user behaviors | |
| Bayram et al. | Could petrol stations play a key role in transportation electrification? a gis-based coverage maximization of fast ev chargers in urban environment | |
| CN114418300A (en) | Multi-type electric vehicle charging facility planning method based on urban function partition and resident trip big data | |
| CN116362523B (en) | Coordinated optimization method for site selection and operation strategy of power exchange station considering temperature adaptability | |
| CN115759779A (en) | Electric vehicle charging station site selection method, electronic equipment and storage medium | |
| CN113343358B (en) | Electric vehicle charging load space-time distribution modeling method considering road information | |
| CN111199080A (en) | Urban electric vehicle charging facility planning method considering different types of charging requirements | |
| CN115239004A (en) | Charging parking lot location and volume optimization method based on charging demand prediction |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200110 |