CN105760949A - Optimizing configuration method for amount of chargers of electromobile charging station - Google Patents

Abstract

The invention relates to an optimizing configuration method for the amount of chargers of an electromobile charging station. The method comprises the following steps that (1) traffic flow of a selected charging station position is investigated, and the distribution relation between the traffic flow and time is analyzed; (2) a distribution rule of charging vehicles is extracted according to traffic flow information, and charging requirement of the fast charging station is calculated; (3) the mechanism service cost in unit time is calculated according to the construction cost, operation and maintenance cost, network loss cost and operation time limit of the fast charging station; (4) travel time of users of a city is used as waiting cost of the users; and (5) aimed at minimizing the expected value of the average total cost per day, the optimal amount of chargers of the charging station is calculated via a random service system model. Waiting cost and operation cost of the fast charging station are taken into full consideration during design, capacity configuration of the fast charging station can meet charging requirements of users, the service level is ensured, resources are optimized in configuration, and unnecessary waste is avoided.

Description

A kind of electric automobile fills station charger number of units Optimal Configuration Method soon

Technical field

The present invention relates to electric automobile charging station planning field, especially a kind of electric automobile fills station charger number of units Optimal Configuration Method soon.

Background technology

Along with economic fast development, the problem of scarcity of resources and environmental pollution is increasingly severe, and automobile is as the walking-replacing tool of people, is deep into every family gradually, and total vehicle also rises at high speed, and this makes resource and environmental problem more worsen.And electric automobile receives the concern of many countries and enterprise due to the characteristic of its energy-conserving and environment-protective.At present, the research and development of electric automobile correlation technique carry out in high gear, industrialization and business-like pattern also basic forming, and also some a considerable number of product formally puts into operation.Along with national governments' support in fund and policy, in predictable future, the application of electric automobile will more and more extensively be changed.

At present, electric automobile is in Demonstration Application stage, officer's car and bus more and accounts for relatively larger, and private car accounts for smaller, electric automobile is filled soon to the planning at station with layout also in research and exploratory stage.The traffic route of officer's car and bus is comparatively fixing with the charging interval, and charge requirement is relatively stable, so the charging station towards officer's car and bus mostly is trickle charge station, the many quantity according to officer's car and bus of capacity configuration configures.And owing to the randomness of private car charging is relatively big, filling station soon capacity configuration be more difficult and determine towards private car charge requirement.If configuring charger number of units by the charge requirement meeting wagon flow peak period, then easily causing the charger most of the time is in idle state, and utilization rate is not high；If by the wagon flow demand of off-peak period configuration charger number of units, then easily causing blocking up of peak period, service level is not high.So, what how not only to meet the basic charge requirement of private car user but also realize charger efficiently utilizes the key being to fill station capacity configuration at present soon.

Summary of the invention

It is an object of the invention to provide a kind of electric automobile and fill station charger number of units Optimal Configuration Method soon, it is ensured that the service level to user, achieve again distributing rationally of resource, it is to avoid unnecessary waste.

For achieving the above object, the present invention adopts following technical proposals:

A kind of electric automobile fills station charger number of units Optimal Configuration Method soon, comprises the following steps:

(1) the wagon flow situation of station addressing place is filled in investigation soon, analyzes the distribution relation of wagon flow and time；

(2) regularity of distribution according to the abstract charging vehicle of car flow information, the charge requirement at station is filled in measuring and calculating soon；

(3) according to filling the construction cost at station, operation and maintenance cost, cost of losses and the Institution Services cost of operation time limit measuring and calculating unit interval soon；

(4) using the user's travel time cost in this city delay cost as user, travel time cost is user's average queue length；

(5) Theories Model of Random Service System is utilized, minimum for target with the expected value of average one day total cost, within average one day, total cost includes Institution Services cost and user's delay cost sum, with Institution Services cost, user's delay cost, charging vehicle charge requirement for input, measuring and calculating soon fill station charger optimum number of units configuration.

The wagon flow situation of station addressing place is filled in described investigation soon, and the distribution relation of analysis wagon flow and time is specially investigation and fills the wagon flow situation of station addressing place soon, was divided into m time period by one day 24 hours, is designated as t₁, t₂... t_i... t_m, then the vehicle flowrate that each time period is corresponding can be designated as Q₁, Q₂... Q_i... Q_m, then the vehicle flowrate Q of whole day is represented by:Time-constrain is:IfFor the average vehicle flow of each time period, then the time of one day can be divided into three periods by the size of vehicle flowrate: peak period, general period and low-valley interval, the wagon flow total amount of peak period is designated as Q_GF, the time span of peak period is designated as t_GF；The wagon flow total amount of general period is designated as Q_YB, the time span of general period is designated as t_YB；The wagon flow total amount of low-valley interval is designated as Q_DG, the time span of low-valley interval is designated as t_DG；Its wagon flow total amount and time span are represented by:

$\left\{\begin{array}{c}{Q}_{GF}=\&Sigma;Q_i,t_{GF}=\&Sigma;t_i,(Q_i>\frac{2}{3}\stackrel{\&OverBar;}{Q})\\ Q_{YB}=\&Sigma;Q_i,t_{YB}=\&Sigma;t_i,(\frac{1}{3}\stackrel{\&OverBar;}{Q}\&le;Q_i\&le;\frac{2}{3}\stackrel{\&OverBar;}{Q})\\ Q_{DG}=\&Sigma;Q_i,t_{DG}=\&Sigma;t_i,(Q_i<\frac{1}{3}\stackrel{\&OverBar;}{Q})\end{array}\right..$

The described regularity of distribution according to the abstract charging vehicle of car flow information, measuring and calculating is filled the charge requirement at station soon and is comprised the following steps:

A, try to achieve peak period, general period and three unit interval time period of low-valley interval and on average arrive and fill station soon and be charged the number of users λ of service, be respectively as follows:

$\left\{\begin{array}{c}{\mathrm{\&lambda;}}_{GF}=\frac{Q_{GF}\&CenterDot;\mathrm{\&alpha;}\&CenterDot;\mathrm{\&beta;}}{t_{GF}}\\ {\mathrm{\&lambda;}}_{YB}=\frac{Q_{YB}\&CenterDot;\mathrm{\&alpha;}\&CenterDot;\mathrm{\&beta;}}{t_{YB}}\\ {\mathrm{\&lambda;}}_{DG}=\frac{Q_{DG}\&CenterDot;\mathrm{\&alpha;}\&CenterDot;\mathrm{\&beta;}}{t_{DG}}\end{array}\right.$

In formula, α is electric automobile quantity accounting in total vehicle in wagon flow, and β is the charge rate of electric automobile, namely needs to carry out the ratio of quick charge in all electric automobiles；

B, set up and fill the standard Multiple server stations service system model immediately at station soon: it is that to obey parameter be the Poisson distribution of λ that user arrives the interval of stochastic service system, λ is the quantity that in the unit time, user on average arrives service system, if μ user of service in the facility unit interval, then the service time of each information desk is submit to the quantum condition entropy that parameter is μ, the work of each information desk simultaneously is independent from, namely do not do cooperation between each information desk simultaneously, then when n>=c, the average service rate of whole service organization is c μ, at n<during c, the average service rate of whole service organization is n μ, the average utilization of service organizationWith P_nT () represents the probability having n user in t service system, remember P_nT () is P_n, then the systematic steady state equilibrium equation of the standard Multiple server stations stochastic service system filling station soon is:

$\left\{\begin{array}{c}{\mathrm{\&mu;P}}_1={\mathrm{\&lambda;P}}_0\\ (n+1){\mathrm{\&mu;P}}_{n+1}+{\mathrm{\&lambda;P}}_{n-1}=(\mathrm{\&lambda;}+n\mathrm{\&mu;})P_n,(1\&le;n\&le;c)\\ {\mathrm{c\&mu;P}}_{n+1}+{\mathrm{\&lambda;P}}_{n-1}=(\mathrm{\&lambda;}+n\mathrm{\&mu;})P_n,(n>c)\end{array}\right.$

Wherein,AndBeing obtained system mode probability by recurrence relation is:

$\left\{\begin{array}{c}{P}_0={\&lsqb;\underset{k=0}{\overset{c-1}{\&Sigma;}}\frac{1}{k!}{\left(\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}\right)}^k+\frac{1}{c!}\&CenterDot;\frac{1}{1-\mathrm{\&rho;}}\&CenterDot;{\left(\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}\right)}^c\&rsqb;}^{-1}\\ P_n=\left\{\begin{array}{c}\frac{1}{n!}{\left(\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}\right)}^n\&CenterDot;P_0,(n\&le;c)\\ \frac{1}{c!\&CenterDot;c^{n-c}}{\left(\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}\right)}^n\&CenterDot;P_0,(n>c)\end{array}\right.\end{array}\right..$

Calculating user's average queue length step is:

A, calculating do not accept the number of users of service and average queue length L in queue_q,

$L_q=\underset{n=c+1}{\overset{\mathrm{\&infin;}}{\&Sigma;}}(n-c)P_H=\underset{k=1}{\overset{\mathrm{\&infin;}}{\&Sigma;}}{\mathrm{kP}}_{k+c}=\underset{k=1}{\overset{\mathrm{\&infin;}}{\&Sigma;}}\frac{k}{c!\&CenterDot;c^k}{\left(c\mathrm{\&rho;}\right)}^{k+c}{P}_0=\frac{{\left(c\mathrm{\&rho;}\right)}^c\&CenterDot;\mathrm{\&rho;}}{c!{(1-\mathrm{\&rho;})}^2}\&CenterDot;P_0,$

In formula, $c\mathrm{\&rho;}=\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}},k=n-c;$

The length of queue and average queue length L in b, computing system_s,

$L_s=L_q+\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}=\frac{{\left(c\mathrm{\&rho;}\right)}^c\&CenterDot;\mathrm{\&rho;}}{c!{(1-\mathrm{\&rho;})}^2}\&CenterDot;P_0+\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}$

Then peak period, general period and three time periods of low-valley interval fill service system queue length in station soon and are respectively as follows:

$\left\{\begin{array}{c}{L}_s({\mathrm{\&lambda;}}_{GF},c)=\frac{{\left({\mathrm{c\&rho;}}_{GF}\right)}^c\&CenterDot;{\mathrm{\&rho;}}_{GF}}{c!{(1-{\mathrm{\&rho;}}_{GF})}^2}\&CenterDot;P_0+\frac{{\mathrm{\&lambda;}}_{GF}}{\mathrm{\&mu;}},({\mathrm{\&rho;}}_{GF}=\frac{{\mathrm{\&lambda;}}_{GF}}{c\mathrm{\&mu;}})\\ L_s({\mathrm{\&lambda;}}_{YB},c)=\frac{{\left({\mathrm{c\&rho;}}_{YB}\right)}^c\&CenterDot;{\mathrm{\&rho;}}_{YB}}{c!{(1-{\mathrm{\&rho;}}_{YB})}^2}\&CenterDot;P_0+\frac{{\mathrm{\&lambda;}}_{YB}}{\mathrm{\&mu;}},({\mathrm{\&rho;}}_{YB}=\frac{{\mathrm{\&lambda;}}_{YB}}{c\mathrm{\&mu;}})\\ L_s({\mathrm{\&lambda;}}_{DG},c)=\frac{{\left({\mathrm{c\&rho;}}_{DG}\right)}^c\&CenterDot;{\mathrm{\&rho;}}_{DG}}{c!{(1-{\mathrm{\&rho;}}_{DG})}^2}\&CenterDot;P_0+\frac{{\mathrm{\&lambda;}}_{DG}}{\mathrm{\&mu;}},({\mathrm{\&rho;}}_{DG}=\frac{{\mathrm{\&lambda;}}_{DG}}{c\mathrm{\&mu;}})\end{array}\right..$

The total cost Z of average one day is Institution Services cost and user's delay cost sum, and the expected value of the total cost Z of average a day is:

Z=24 c_s·c+c_w·L_s(λ_GF, c) t_GF+c_w·L_s(λ_YB, c) t_YB+c_w·L_s(λ_DC, C) and t_DG=24 c_s·c+c_w·[L_s(λ_GF, c) t_GF+L_s(λ_YB, c) t_YB+L_s(λ_DG, c) t_DG]

Wherein, c_sFor average each charger cost of serving hourly, c_wFor each user fill soon station in waiting cost hourly, L_sIt is the function of c, remembers c^*For the optimal allocation number of units of charger, then Z (c^*) it is least cost, adopt technique of marginal analysis to solve, then have:

$\left\{\begin{array}{c}Z\left(c^{*}\right)\&le;Z(c^{*}-1)\\ Z\left(c^{*}\right)\&le;Z(c^{*}+1)\end{array}\right.$

The optimum charger configuration number of units c filling station soon can be tried to achieve^*。

The invention has the beneficial effects as follows,

The delay cost of user is included in the consideration category filling station capacity configuration soon by the present invention, utilizes the change of user's delay cost to reflect the service level filling station soon, considers operation cost more objectivity than simple.Utilize theory of random service system in the advantage of Service System Optimization design aspect, structure electric automobile fills the optimum number of units of station charger soon and designs a model, for filling station soon, if charger configuration is more, the cost of serving then filling station soon can increase, service level can increase, and user is filling in station soon owing to waiting that the cost spent can reduce；On the contrary, if charger configuration is less, the cost of serving filling station soon can reduce, and service level can reduce, and user is filling in station soon owing to waiting that the cost spent can increase.Take into full account during design that user is filling the delay cost at station soon and filling the operation cost at station soon, so that the capacity configuration filling station soon had both met the charge requirement of user, utilize theory of random service system by user fill soon station delay cost and fill soon station operation cost this paradox organically combined consider, make to fill soon to stand in and ensure that service level and equipment high efficiency utilize two aspects to reach optimal balance, it is achieved that maximizing the benefits.Ensure that the service level to user, achieve again distributing rationally of resource, it is to avoid unnecessary waste.

Accompanying drawing explanation

Fig. 1 is the general structure model filling station service system soon；

Fig. 2 fills station charger number of units to distribute workflow rationally soon；

Fig. 3 is the systematic state transfer graph of a relation of M/M/c model；

Fig. 4 is system cost model schematic.

Detailed description of the invention

As shown in Figure 1, the service process filling station soon is usually: the electric automobile user serviced totally enters from user and fills station service system (i.e. input process) soon, waiting in line to be serviced before arriving service organization, acceptance service is left (i.e. output procedure) after completing immediately.

As in figure 2 it is shown, a kind of electric automobile fills station charger number of units Optimal Configuration Method soon, comprise the following steps:

(1) the wagon flow situation of station addressing place is filled in investigation soon, analyzes the distribution relation of wagon flow and time；

(2) regularity of distribution according to the abstract charging vehicle of car flow information, the charge requirement at station is filled in measuring and calculating soon；

(3) according to filling the construction cost at station, operation and maintenance cost, cost of losses and the Institution Services cost of operation time limit measuring and calculating unit interval soon；

(4) using the user's travel time cost in this city delay cost as user, travel time cost is user's average queue length；

(5) Theories Model of Random Service System is utilized, minimum for target with the expected value of average one day total cost (Institution Services cost and user's delay cost sum), with Institution Services cost, user's delay cost, charging vehicle charge requirement for input, measuring and calculating soon fill station charger optimum number of units configuration.

The wagon flow situation of station addressing place is filled in described investigation soon, and the distribution relation of analysis wagon flow and time is specially investigation and fills the wagon flow situation of station addressing place soon, was divided into m time period by one day 24 hours, is designated as t₁, t₂... t_i... t_m, then the vehicle flowrate that each time period is corresponding can be designated as Q1_,Q₂... Q_i... Q_m,

Then the vehicle flowrate Q of whole day is represented by:

Time-constrain is: $\underset{i=1}{\overset{m}{\&Sigma;}}{t}_i=24,$

IfFor the average vehicle flow of each time period,

Then the time of one day can be divided into three periods by the size of vehicle flowrate: peak period, (wagon flow total amount was designated as: Q_GF, time span is designated as: t_GF), its wagon flow total amount and time span are represented by:

$\left\{\begin{array}{c}{Q}_{GF}=\&Sigma;Q_i,t_{GF}=\&Sigma;t_i,(Q_i>\frac{2}{3}\stackrel{\&OverBar;}{Q})\\ Q_{YB}=\&Sigma;Q_i,t_{YB}=\&Sigma;t_i,(\frac{1}{3}\stackrel{\&OverBar;}{Q}\&le;Q_i\&le;\frac{2}{3}\stackrel{\&OverBar;}{Q})\\ Q_{DG}=\&Sigma;Q_i,t_{DG}=\&Sigma;t_i,(Q_i<\frac{1}{3}\stackrel{\&OverBar;}{Q})\end{array}\right..$

The described regularity of distribution according to the abstract charging vehicle of car flow information, measuring and calculating is filled the charge requirement at station soon and is comprised the following steps:

A, try to achieve peak period, general period and three unit interval time period of low-valley interval and on average arrive and fill station soon and be charged the number of users λ of service, be respectively as follows:

$\left\{\begin{array}{c}{\mathrm{\&lambda;}}_{GF}=\frac{Q_{GF}\&CenterDot;\mathrm{\&alpha;}\&CenterDot;\mathrm{\&beta;}}{t_{GF}}\\ {\mathrm{\&lambda;}}_{YB}=\frac{Q_{YB}\&CenterDot;\mathrm{\&alpha;}\&CenterDot;\mathrm{\&beta;}}{t_{YB}}\\ {\mathrm{\&lambda;}}_{DG}=\frac{Q_{DG}\&CenterDot;\mathrm{\&alpha;}\&CenterDot;\mathrm{\&beta;}}{t_{DG}}\end{array}\right.$

In formula, α is electric automobile quantity accounting in total vehicle in wagon flow, and β is the charge rate of electric automobile, namely needs to carry out the ratio of quick charge in all electric automobiles；

B, set up and fill the standard Multiple server stations service system model immediately at station soon；

In Theories Model of Random Service System, topmost three factors are:

1) user arrives the regularity of distribution of the interval of system in succession；

2) Time Distribution that user is serviced in systems；

3) quantity of information desk in stochastic service system.

According to three above factor classification, then the general type of Theories Model of Random Service System is: X/Y/Z/A/B/C

Wherein, X represents that user arrives the regularity of distribution of the interval of stochastic service system in succession；Y represents the Time Distribution that user is serviced in stochastic service system；Z represents the quantity of information desk in stochastic service system；A represents the capacity of system；B represents the quantity of user source；C represents service regulation (being generally defaulted as First Come First Served).

Represent that the interval in succession arrived distribution X and numbers distribution in system Y has following several situation:

1) M (Markou) represents quantum condition entropy；

2) D (Deterministic) represents that deterministic type is distributed；

3) Ek (Erlang) represents that k rank Ai Erlang is distributed；

4) GI (GeneralIndependent) represents generally separate interval distribution；

5) G (General) represents that general service-time is distributed.

As it is shown on figure 3, the Theories Model of Random Service System of Multiple server stations is exactly comprise multiple separate information desk in service organization.Three conditions of Poisson flow are met owing to user arrives service system, i.e. markov property, stationarity and universality, user arrive the interval of stochastic service system be obey parameter be the Poisson distribution of λ, λ is the quantity that in the unit time, user on average arrives service system, if μ user of service in the facility unit interval, then the service time of each information desk is submit to the quantum condition entropy that parameter is μ, not (not the doing cooperation) that the work of each information desk simultaneously is independent from, then the average service rate of whole service organization is c μ (as n>=c) or n μ (as n<c), the average utilization of service organizationWith P_nT () represents the probability having n user in t service system, remember P_nT () is P_n, then the systematic steady state equilibrium equation of the standard Multiple server stations stochastic service system filling station soon is:

$\left\{\begin{array}{c}{\mathrm{\&mu;P}}_1={\mathrm{\&lambda;P}}_0\\ (n+1){\mathrm{\&mu;P}}_{n+1}+{\mathrm{\&lambda;P}}_{n-1}=(\mathrm{\&lambda;}+n\mathrm{\&mu;})P_n,(1\&le;n\&le;c)\\ {\mathrm{c\&mu;P}}_{n+1}+{\mathrm{\&lambda;P}}_{n-1}=(\mathrm{\&lambda;}+n\mathrm{\&mu;})P_n,(n>c)\end{array}\right.$

Wherein,AndBeing obtained system mode probability by recurrence relation is:

$\left\{\begin{array}{c}{P}_0={\&lsqb;\underset{k=0}{\overset{c-1}{\&Sigma;}}\frac{1}{k!}{\left(\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}\right)}^k+\frac{1}{c!}\&CenterDot;\frac{1}{1-\mathrm{\&rho;}}\&CenterDot;{\left(\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}\right)}^c\&rsqb;}^{-1}\\ P_n=\left\{\begin{array}{c}\frac{1}{n!}{\left(\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}\right)}^n\&CenterDot;P_0,(n\&le;c)\\ \frac{1}{c!\&CenterDot;c^{n-c}}{\left(\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}\right)}^n\&CenterDot;P_0,(n>c)\end{array}\right.\end{array}\right..$

Calculating user's average queue length step is:

A, calculate average queue length L_q(in queue, namely not accepting the number of users of service)

$L_q=\underset{n=c+1}{\overset{\mathrm{\&infin;}}{\&Sigma;}}(n-c)P_n=\underset{k=1}{\overset{\mathrm{\&infin;}}{\&Sigma;}}{\mathrm{kP}}_{k+c}=\underset{k=1}{\overset{\mathrm{\&infin;}}{\&Sigma;}}\frac{k}{c!\&CenterDot;c^k}{\left(c\mathrm{\&rho;}\right)}^{k+c}{P}_0=\frac{{\left(c\mathrm{\&rho;}\right)}^c\&CenterDot;\mathrm{\&rho;}}{c!{(1-\mathrm{\&rho;})}^2}\&CenterDot;P_0,$

In formula, $c\mathrm{\&rho;}=\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}},k=n-c;$

B, calculating average queue length L_s(i.e. the length of queue in system)

$L_s=L_q+\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}=\frac{{\left(c\mathrm{\&rho;}\right)}^c\&CenterDot;\mathrm{\&rho;}}{c!{(1-\mathrm{\&rho;})}^2}\&CenterDot;P_0+\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}$

Then peak period, general period and three time periods of low-valley interval fill service system queue length in station soon and are respectively as follows:

$\left\{\begin{array}{c}{L}_s({\mathrm{\&lambda;}}_{GF},c)=\frac{{\left({\mathrm{c\&rho;}}_{GF}\right)}^c\&CenterDot;{\mathrm{\&rho;}}_{GF}}{c!{(1-{\mathrm{\&rho;}}_{GF})}^2}\&CenterDot;P_0+\frac{{\mathrm{\&lambda;}}_{GF}}{\mathrm{\&mu;}},({\mathrm{\&rho;}}_{GF}=\frac{{\mathrm{\&lambda;}}_{GF}}{c\mathrm{\&mu;}})\\ L_s({\mathrm{\&lambda;}}_{YB},c)=\frac{{\left({\mathrm{c\&rho;}}_{YB}\right)}^c\&CenterDot;{\mathrm{\&rho;}}_{YB}}{c!{(1-{\mathrm{\&rho;}}_{YB})}^2}\&CenterDot;P_0+\frac{{\mathrm{\&lambda;}}_{YB}}{\mathrm{\&mu;}},({\mathrm{\&rho;}}_{YB}=\frac{{\mathrm{\&lambda;}}_{YB}}{c\mathrm{\&mu;}})\\ L_s({\mathrm{\&lambda;}}_{DG},c)=\frac{{\left({\mathrm{c\&rho;}}_{DG}\right)}^c\&CenterDot;{\mathrm{\&rho;}}_{DG}}{c!{(1-{\mathrm{\&rho;}}_{DG})}^2}\&CenterDot;P_0+\frac{{\mathrm{\&lambda;}}_{DG}}{\mathrm{\&mu;}},({\mathrm{\&rho;}}_{DG}=\frac{{\mathrm{\&lambda;}}_{DG}}{c\mathrm{\&mu;}})\end{array}\right..$

As shown in Figure 4, for filling station soon, if charger configuration is more, then the cost of serving filling station soon can increase, and service level can increase, and user is filling in station soon owing to waiting that the cost spent can reduce；On the contrary, if charger configuration is less, the cost of serving filling station soon can reduce, and service level can reduce, and user is filling in station soon owing to waiting that the cost spent can increase.Both expenses are conflict bodies relative to the charger number of units filling station soon, but both sums then there will be a minima when the charger number of units filling station soon reaches a certain value.So, this motion has taken into account the interests filling station soon with user both sides, distributes, so that the delay cost sum of the cost of serving filling station soon and user is minimum, the charger number of units filling station soon rationally for target.

Calculating charger cost of serving step is:

If building k quick charge station, wherein, each quick charge station configuration z_iPlatform charger, actual according to engineering, construction cost and cost of losses sum are C_T, then:

$C_T=\mathrm{\&gamma;}\&CenterDot;\underset{i=1}{\overset{k}{\&Sigma;}}(C^w\&CenterDot;z_i+C^Y\&CenterDot;z_i+C^F)$

Wherein γ is recovery of the capital coefficient,ε is Annual Percentage Rate, and T is the length of service filling station soon, C^wPart component relevant with filling geographical position, station soon in variable cost, C is related to for filling station soon^YPart component unrelated with filling geographical position, station soon in variable cost, C is related to for filling station soon^FFor filling the fixed cost that station is built, z soon_iThe quantity filling machine soon that station is equipped with is filled soon for each.

If average each charger cost of serving hourly is c_s, then

$c_s=\frac{C_T}{8760\&CenterDot;\underset{i=1}{\overset{k}{\&Sigma;}}{z}_i}$

If each user is c filling waiting cost hourly in station soon_w, then the expected value of the total cost Z (Institution Services cost and user's delay cost sum) of average a day is:

Z=24 c_s·c+c_w·L_s(λ_GF, c) t_GF+c_w·L_s(λ_YB, c) t_YB+c_w·L_s(λ_DC, C) and t_DG=24 c_s·c+c_w·[L_s(λ_GF, c) t_GF+L_s(λ_YB, c) t_YB+L_s(λ_DG, c) t_DG]

Wherein, L_sIt is the function of c, remembers c^*For the optimal allocation number of units of charger, then Z (c^*) it is least cost, adopt technique of marginal analysis to solve, then have:

$\left\{\begin{array}{c}Z\left(c^{*}\right)\&le;Z(c^{*}-1)\\ Z\left(c^{*}\right)\&le;Z(c^{*}+1)\end{array}\right.$

The optimum charger configuration number of units c filling station soon can be tried to achieve^*。

The specific embodiment of the present invention is described in conjunction with accompanying drawing although above-mentioned; but not limiting the scope of the invention; one of ordinary skill in the art should be understood that; on the basis of technical scheme, those skilled in the art need not pay various amendments or deformation that creative work can make still within protection scope of the present invention.

Claims (5)

1. electric automobile fills a station charger number of units Optimal Configuration Method soon, it is characterized in that, comprises the following steps:

(1) the wagon flow situation of station addressing place is filled in investigation soon, analyzes the distribution relation of wagon flow and time；

(2) regularity of distribution according to the abstract charging vehicle of car flow information, the charge requirement at station is filled in measuring and calculating soon；

(3) according to filling the construction cost at station, operation and maintenance cost, cost of losses and the Institution Services cost of operation time limit measuring and calculating unit interval soon；

(4) using the user's travel time cost in this city delay cost as user, travel time cost is user's average queue length；

(5) Theories Model of Random Service System is utilized, minimum for target with the expected value of average one day total cost, within average one day, total cost includes Institution Services cost and user's delay cost sum, with Institution Services cost, user's delay cost, charging vehicle charge requirement for input, measuring and calculating soon fill station charger optimum number of units configuration.

2. a kind of electric automobile as claimed in claim 1 fills station charger number of units Optimal Configuration Method soon, it is characterized in that, the wagon flow situation of station addressing place is filled in described investigation soon, the distribution relation of analysis wagon flow and time is specially investigation and fills the wagon flow situation of station addressing place soon, it is divided into m time period by one day 24 hours, is designated as t₁, t₂... t_i... t_m, then the vehicle flowrate that each time period is corresponding can be designated as Q₁, Q₂... Q_i... Q_m, then the vehicle flowrate Q of whole day is represented by:Time-constrain is:IfFor the average vehicle flow of each time period, then the time of one day can be divided into three periods by the size of vehicle flowrate: peak period, general period and low-valley interval, the wagon flow total amount of peak period is designated as Q_GF, the time span of peak period is designated as t_GF；The wagon flow total amount of general period is designated as Q_YB, the time span of general period is designated as t_YB；The wagon flow total amount of low-valley interval is designated as Q_DG, the time span of low-valley interval is designated as t_DG；Its wagon flow total amount and time span are represented by:

$\left\{\begin{array}{c}{Q}_{GF}={\mathrm{\&Sigma;Q}}_i,t_{GF}={\mathrm{\&Sigma;t}}_i,(Q_i>\frac{2}{3}\stackrel{\&OverBar;}{Q})\\ Q_{YB}={\mathrm{\&Sigma;Q}}_i,t_{YB}={\mathrm{\&Sigma;t}}_i,(\frac{1}{3}\stackrel{\&OverBar;}{Q}\&le;Q_i\&le;\frac{2}{3}\stackrel{\&OverBar;}{Q})\\ Q_{DG}={\mathrm{\&Sigma;Q}}_i,t_{DG}={\mathrm{\&Sigma;t}}_i,(Q_i<\frac{1}{3}\stackrel{\&OverBar;}{Q})\end{array}\right..$

3. a kind of electric automobile as claimed in claim 1 fills station charger number of units Optimal Configuration Method soon, it is characterized in that, the described regularity of distribution according to the abstract charging vehicle of car flow information, and measuring and calculating is filled the charge requirement at station soon and comprised the following steps:

A, try to achieve peak period, general period and three unit interval time period of low-valley interval and on average arrive and fill station soon and be charged the number of users λ of service, be respectively as follows:

$\left\{\begin{array}{c}{\mathrm{\&lambda;}}_{GF}=\frac{Q_{GF}\&CenterDot;\mathrm{\&alpha;}\&CenterDot;\mathrm{\&beta;}}{t_{GF}}\\ {\mathrm{\&lambda;}}_{YB}=\frac{Q_{YB}\&CenterDot;\mathrm{\&alpha;}\&CenterDot;\mathrm{\&beta;}}{t_{YB}}\\ {\mathrm{\&lambda;}}_{DG}=\frac{Q_{DG}\&CenterDot;\mathrm{\&alpha;}\&CenterDot;\mathrm{\&beta;}}{t_{DG}}\end{array}\right.$

In formula, α is electric automobile quantity accounting in total vehicle in wagon flow, and β is the charge rate of electric automobile, namely needs to carry out the ratio of quick charge in all electric automobiles；

B, set up and fill the standard Multiple server stations service system model immediately at station soon: it is that to obey parameter be the Poisson distribution of λ that user arrives the interval of stochastic service system, λ is the quantity that in the unit time, user on average arrives service system, if μ user of service in the facility unit interval, then the service time of each information desk is submit to the quantum condition entropy that parameter is μ, the work of each information desk simultaneously is independent from, namely do not do cooperation between each information desk simultaneously, then when n>=c, the average service rate of whole service organization is c μ, at n<during c, the average service rate of whole service organization is n μ, the average utilization of service organizationWith P_nT () represents the probability having n user in t service system, remember P_nT () is P_n, then the systematic steady state equilibrium equation of the standard Multiple server stations stochastic service system filling station soon is:

$\left\{\begin{array}{c}\mathrm{\&mu;}{P}_1=\mathrm{\&lambda;}{P}_0\\ (n+1){\mathrm{\&mu;P}}_{n+1}+{\mathrm{\&lambda;P}}_{n-1}=(\mathrm{\&lambda;}+n\mathrm{\&mu;})P_n,(1\&le;n\&le;c)\\ {\mathrm{c\&mu;P}}_{n+1}+{\mathrm{\&lambda;P}}_{n+1}=(\mathrm{\&lambda;}+n\mathrm{\&mu;})P_n,(n>c)\end{array}\right.$

Wherein,AndBeing obtained system mode probability by recurrence relation is:

$\{\begin{array}{c}{P}_0={\&lsqb;\underset{k=0}{\overset{c-1}{\mathrm{\&Sigma;}}}\frac{1}{k!}{\left(\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}\right)}^k+\frac{1}{c!}\&CenterDot;\frac{1}{1-\mathrm{\&rho;}}\&CenterDot;{\left(\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}\right)}^c\&rsqb;}^{-1}\\ P_n=\left\{\begin{array}{c}\frac{1}{n!}{\left(\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}\right)}^n\&CenterDot;P_n,(n\&le;c)\\ \frac{1}{c!\&CenterDot;c^{n-c}}{\left(\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}\right)}^n\&CenterDot;P_0,(n>c)\end{array}\right.\end{array}.$

4. a kind of electric automobile as claimed in claim 1 fills station charger number of units Optimal Configuration Method soon, it is characterized in that, calculates userAverage queue lengthStep is:

A, calculating do not accept the number of users of service and average queue length L in queue_q,

$L_q=\underset{n=c+1}{\overset{\mathrm{\&infin;}}{\mathrm{\&Sigma;}}}(n-c)P_n=\underset{k=1}{\overset{\mathrm{\&infin;}}{\&Sigma;}}{\mathrm{kP}}_{k+c}=\underset{k=1}{\overset{\mathrm{\&infin;}}{\&Sigma;}}\frac{k}{c!\&CenterDot;c^k}{\left(c\mathrm{\&rho;}\right)}^{k+c}{P}_0=\frac{{\left(c\mathrm{\&rho;}\right)}^c\&CenterDot;\mathrm{\&rho;}}{c!{(1-\mathrm{\&rho;})}^2}\&CenterDot;P_0,$

In formula, $c\mathrm{\&rho;}=\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}},k=n-c;$

The length of queue and average queue length L in b, computing system_s,

$L_s=L_q+\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}=\frac{{\left(c\mathrm{\&rho;}\right)}^c\&CenterDot;\mathrm{\&rho;}}{c!{(1-\mathrm{\&rho;})}^2}\&CenterDot;P_0+\frac{\mathrm{\&lambda;}}{\mathrm{\&mu;}}$

Then peak period, general period and three time periods of low-valley interval fill service system queue length in station soon and are respectively as follows:

$\left\{\begin{array}{c}{L}_s({\mathrm{\&lambda;}}_{GF},c)=\frac{{\left({\mathrm{c\&rho;}}_{GF}\right)}^c\&CenterDot;{\mathrm{\&rho;}}_{GF}}{c!{(1-{\mathrm{\&rho;}}_{GF})}^2}\&CenterDot;P_0+\frac{{\mathrm{\&lambda;}}_{GF}}{\mathrm{\&mu;}},({\mathrm{\&rho;}}_{GF}=\frac{{\mathrm{\&lambda;}}_{GF}}{c\mathrm{\&mu;}})\\ L_s({\mathrm{\&lambda;}}_{YB},c)=\frac{{\left({\mathrm{c\&rho;}}_{YB}\right)}^c\&CenterDot;{\mathrm{\&rho;}}_{YB}}{c!{(1-{\mathrm{\&rho;}}_{YB})}^2}\&CenterDot;P_0+\frac{{\mathrm{\&lambda;}}_{YB}}{\mathrm{\&mu;}},({\mathrm{\&rho;}}_{YB}=\frac{{\mathrm{\&lambda;}}_{YB}}{c\mathrm{\&mu;}})\\ L_s({\mathrm{\&lambda;}}_{DG},c)=\frac{{\left({\mathrm{c\&rho;}}_{DG}\right)}^c\&CenterDot;{\mathrm{\&rho;}}_{DG}}{c!{(1-{\mathrm{\&rho;}}_{DG})}^2}\&CenterDot;P_0+\frac{{\mathrm{\&lambda;}}_{DG}}{\mathrm{\&mu;}},({\mathrm{\&rho;}}_{DG}=\frac{{\mathrm{\&lambda;}}_{DG}}{c\mathrm{\&mu;}})\end{array}\right..$

5. a kind of electric automobile as claimed in claim 1 fills station charger number of units Optimal Configuration Method soon, it is characterized in that, the total cost Z of average a day is Institution Services cost and user's delay cost sum, and the expected value of the total cost Z of average a day is:

$\begin{array}{c}Z=24\&CenterDot;c_s\&CenterDot;c+c_w\&CenterDot;L_s({\mathrm{\&lambda;}}_{GF},c)\&CenterDot;t_{GF}+c_w\&CenterDot;L_s({\mathrm{\&lambda;}}_{YB},c)\&CenterDot;t_{YB}+c_w\&CenterDot;L_s({\mathrm{\&lambda;}}_{DG},c)\&CenterDot;t_{DG}\\ =24\&CenterDot;c_s\&CenterDot;c+c_w\&CenterDot;\&lsqb;L_s({\mathrm{\&lambda;}}_{GF},c)\&CenterDot;t_{GF}+L_s({\mathrm{\&lambda;}}_{YB},c)\&CenterDot;t_{YB}+L_s({\mathrm{\&lambda;}}_{DG},c)\&CenterDot;t_{DG}\&rsqb;\end{array}$

Wherein, c_sFor average each charger cost of serving hourly, c_wFor each user fill soon station in waiting cost hourly, L_sIt is the function of c, remembers c^*For the optimal allocation number of units of charger, then Z (c^*) it is least cost, adopt technique of marginal analysis to solve, then have:

$\left\{\begin{array}{c}Z\left(c^{*}\right)\&le;Z(c^{*}-1)\\ Z\left(c^{*}\right)\&le;Z(c^{*}+1)\end{array}\right.$

The optimum charger configuration number of units c filling station soon can be tried to achieve^*。