CN103646295B - Electric automobile charging and conversion electric network integration dispatching method based on service station universal model - Google Patents

Abstract

The invention discloses a kind of electric automobile charging and conversion electric network integration dispatching method based on service station universal model, set up the universal model in three types service station in charging and conversion electric network, influence each other and the relation restricted between behavior (charge, change electricity, battery allotment) and the state (full set of cells quantity and sky set of cells quantity) in service station in order to characterize；According in each sampling period of service station initially change electricity demand, utilize queueing theory principles simulation to obtain the actual of service station and change electricity quantity, service station is changed the impact of power fully taking into account the charger quantity in service station, user's average arrival rate and the congestion lengths upper limit；Set up the integrated scheduling model of actual charging and conversion electric network operation, by optimizing the charging scheme of set of cells, programs and logistics scheme so that the running cost of network is minimum.

Description

Electric automobile charging and conversion electric network integration dispatching method based on service station universal model

Technical field

The invention belongs to charging electric vehicle field, be specifically related to modeling and the scheduling batteries method of electric automobile charging and conversion electric network under " concentrating charging, unified dispensing " and charging and conversion electric both of which.

Background technology

The power mode that changes based on rentable battery shows obvious advantage in reducing user and initially purchasing car cost, prolongation battery life, quickening charging interval etc., become one of competitive business model of current power development of automobile, specifically comprise " concentrating charging, unified dispensing " and charging and conversion electric both of which.The main energy sources supply mode of " concentration charging, unified dispensing " pattern is concentrated charging station and dispensing station；The main energy sources supply mode of charging and conversion electric pattern is charging station.In charging and conversion electric network, concentrated charging station undertakes large-scale battery charging function, the battery being full of has small-scale charging ability by being distributed to and changes the charging and conversion electric station of Electricity Functional and only possess the dispensing station changing battery functi on, thus realizing the battery supplied to user.In the actual moving process of charging and conversion electric network, relate to all too many levels such as the charging of relevant battery, battery allotment, logistics distribution, and links is closely related.In order to meet the electricity demand of changing of user, guarantee power grid security, stable, economical operation, scientific and reasonable charging and conversion electric network integration traffic control plays vital effect.

Academic circles at present also rarely has about changing the research of charging and conversion electric network operation scheduling under power mode, only has part document and is discussed for charging optimization therein.In general, existing research mainly using charging station as main body, on the research of its charging load and control strategy entirely without the impact considering the factor such as distribution strategy, distribution time comprehensively.Ignore the actual support effect that battery is allocated by battery allotment space-time balance and logistics deployment to maintaining battery supply and demand.

Summary of the invention

It is an object of the invention to overcome the deficiencies in the prior art, a kind of electric automobile charging and conversion electric network integration dispatching method based on service station universal model is provided, by the general modeling to three types service station, become more meticulous and influence each other and the relation restricted between the various actions and the state that characterize service station；Integrated scheduling model by battery charging, allotment and logistics distribution, realize changing the optimization of charging scheme under power mode, battery programs and logistics distribution scheme, meet user change electricity demand premise under reduce network operation cost as far as possible, for the operation based theoretical of charging and conversion electric network.

A kind of electric automobile charging and conversion electric network integration dispatching method based on service station universal model of the present invention, comprises the following steps:

1) three types service station in charging and conversion electric network is set up: the universal model of concentrated charging station, charging and conversion electric station and dispensing station, to characterize the relation between the state (including full set of cells number and empty set of cells number) in service station and behavior (including changing electricity, charging, battery allotment):

$Q_{i,t}^f=\left\{\begin{array}{cc}{Q}_{i,t-1}^f+Q_{i,t-[t\_\mathrm{chg}/T]}^c-Q_{i,t-1}^h-Q_{i,t-1}^d,& Q_{i,t-1}^d\&GreaterEqual;0\\ Q_{i,t-1}^f+Q_{i,t-[t\_\mathrm{chg}/T]}^c-Q_{i,t-1}^h+\underset{j}{\mathrm{\&Sigma;}}{Q}_{i,t-[{\mathrm{dis}}_{\mathrm{ij}}/T]}^d,& Q_{i,t-[{\mathrm{dis}}_{\mathrm{ij}}/T]}^d<0\end{array}\right.---\left(1\right)$

$Q_{i,t}^e=\left\{\begin{array}{cc}{Q}_{i,t-1}^e+Q_{i,t-1}^h-Q_{i,t-1}^c+\underset{j}{\mathrm{\&Sigma;}}{Q}_{i,t-[{\mathrm{dis}}_{\mathrm{ij}}*2/T]}^d,& Q_{i,t-[{\mathrm{dis}}_{\mathrm{ij}}*2/T]}^d\&GreaterEqual;0\\ Q_{i,t-1}^e+Q_{i,t-1}^h-Q_{i,t-1}^c-\underset{j}{\mathrm{\&Sigma;}}{Q}_{i,t-[{\mathrm{dis}}_{\mathrm{ij}}/T]}^d,& Q_{i,t-[{\mathrm{dis}}_{\mathrm{ij}}/T]}^d<0\end{array}\right.---\left(2\right)$

$\left\{\begin{array}{c}{Q}_{i,t}^h\&le;Q_{i,t}^f\\ Q_{i,t}^c\&le;Q_{i,t}^e\\ Q_{i,t}^d\&le;Q_{i,t}^f-Q_{i,t}^h,\mathrm{if}{Q}_{i,t}^d>0\end{array}\right.---\left(3\right)$

In formula:Full set of cells quantity for t service station i；For the empty set of cells quantity of t service station i, t=1 represents initial time；It is the full set of cells quantity that user changes electricity for t service station i；For the t service station i empty set of cells quantity started to charge up；The full set of cells quantity of allotment is started for t service station i,Represent that t service station i recalls set of cells,Represent that t has full set of cells to recall from other service stations to service station i.T_chg is set of cells charging required time (min)；T is sampling period (min)；dis_ijSet of cells for service station i to service station j allocates minimum one way used time (min)；[] expression rounds up.

2) according to the correlation theory of queueing theory, electric automobile changes electricity service system can regard a queuing system as, and the client of system is electric automobile, and information desk is battery replacement device, and the service that information desk provides is to change battery.Obtain the actual of service station according to queueing theory principle and change power, the original electricity demand curve that changes is modified:

$Q_{i,t}^r=\min (Q_{i,t}^R,N_{i,t})---\left(4\right)$

In formula: N_i,tIn the t sampling period, service station i's changes power；Original for t service station i changes electricity demand；For revise after t service station i change electricity demand.

3) according to network traffic figure, dijkstra algorithm is utilized to solve the shortest used time path；

4) logistics cost runing in a few days charging and conversion electric network is calculated:

$F_d=\underset{t}{\mathrm{\&Sigma;}}\underset{i}{\mathrm{\&Sigma;}}\underset{j}{\mathrm{\&Sigma;}}\mathrm{ceil}(\mathrm{abs}\left(Q_{\mathrm{ij},t}^d\right)/\mathrm{vc})*{\mathrm{dis}}_{\mathrm{ij}}*c_{\mathrm{ev}}/120---\left(5\right)$

In formula: F_dFor logistics cost；The full set of cells number of service station j it is distributed to, if full set of cells is distributed to service station i from service station j, then for t service station iIt is negative；Vc is the set of cells number that each car can load, and unit is block；c_evFor single logistics vehicles freight charges hourly, unit be unit/hour.

5) a charging expense runing in a few days charging and conversion electric network is calculated:

$F_c=\underset{i}{\mathrm{\&Sigma;}}\underset{\mathrm{time}=1}{\overset{48}{\mathrm{\&Sigma;}}}\underset{t=\mathrm{time}}{\overset{\mathrm{time}+[t\_\mathrm{chg}/30]-1}{\mathrm{\&Sigma;}}}{Q}_{i,t}^c*p_{\mathrm{ack}}*T*p_t---\left(6\right)$

In formula: F_cFor charging expense；p_ackFor the charge power of set of cells, unit is KW；p_tFor the electricity price of t, unit is unit.

6) with tou power price for background, with network operation cost minimization for target, actual electric automobile charging and conversion electric network operation integration scheduling model is set up, thus drawing charging and the battery allotment strategy of integration scheduling:

MinF=F_c+F_d(7)

$s.t.Q_{i,t}^h=\min (Q_{i,t}^f,Q_{i,t}^r),\&ForAll;i,t---\left(8\right)$

$Q_{i,t}^h=Q_{i,t}^r,\&ForAll;i,t---\left(9\right)$

$Q_{i,t}^c\&le;\min (Q_{i,t}^e,P_i*1000/p_{\mathrm{ack}}),\&ForAll;i,t---\left(10\right)$

$\underset{j}{\mathrm{\&Sigma;}}\mathrm{ceil}(Q_{\mathrm{ij},t}^d/\mathrm{vc})\&le;{\mathrm{nc}}_i,\mathrm{if}{Q}_{\mathrm{ij},t}^d>0,\&ForAll;i,t---\left(11\right)$

$Q_{i,t}^d\&le;Q_{i,t}^f-Q_{i,t}^h,\mathrm{if}{Q}_{i,t}^d>0,Q_{i,t}^d=\underset{j}{\mathrm{\&Sigma;}}{Q}_{\mathrm{ij},t}^d,\&ForAll;i,t---\left(12\right)$

$Q_{\mathrm{ij},t}^d+Q_{\mathrm{ji},t}^d=0---\left(13\right)$

$\underset{i}{\mathrm{\&Sigma;}}{Q}_{i,t}^d=0,\&ForAll;t---\left(14\right)$

In formula: P_iFor the capacity of service station i, unit is MW；nc_iLogistics vehicles number for website i.

The service station universal model that described step 1 is set up discloses the relation of the restriction mutually that influences each other between the state in service station and behavior:

1) its state of the behavioral implications in service station change

Instant behavior, immediately the state in service station is produced impact as changed electricity, and process state has aftereffect, namely process state can affect service station state sometime thereafter: 1. charging behavior can affect the empty set of cells number in service station when occurring, but will then through the duration that charges, the full set of cells number in service station just can change；2. when battery allotment starts, only impact recalls the service station of full set of cells, set of cells is packed in the service of reception completely set of cells when arriving of allocating and sky set of cells quantity all changes, and until logistics vehicles returns to the service station recalling full set of cells, its sky set of cells number also changes.

2) its behavior of the state constraint in service station

Changing electricity and the battery allotment constraint by full set of cells number, charging is by the constraint of empty set of cells number.

In described step 2, the calculating process of system service ability has been taken into account system service ability and has been subject to the battery replacement device quantity in service station, the average arrival rate of user, acceptable wait to length, the restriction changing the electricity factors such as time, specifically includes following steps:

1) setting up electric automobile and change the M/D/C queuing model of electricity service system, M/D/C represents that quantum condition entropy is obeyed at the client interval time of advent, service time is the model having C information desk in fixed value and system.

2) initially change electricity demand according to service station in each sampling period, calculate its its arrival rate average and the user interval time of advent:

${\mathrm{\&lambda;}}_{i,t}=\frac{Q_{i,t}^R}{T}---\left(15\right)$

$P_{i,t}(T\&le;t)=1-e^{-{\mathrm{\&lambda;}}_{i,t}t}---\left(16\right)$

In formula: λ_i,tIt it is the average arrival rate of service station i in the t sampling period；Be in the t sampling period service station i initially change electricity demand.

3) practical operation situation of matlab analogue simulation certain scale system, it is thus achieved that the electric automobile number of service station active service in each sampling period, calculates service rate η:

4) calculating the actual power that changes in service station is:

N_i,t=λ_i,t*T*η_i,t(18)

The constraints of described step 6 is:

1) what service station provided the user change, and electricity changes the smaller value of electricity demand equal to its full set of cells number and user；

2) system operation meets and basic changes electricity demand；

3) the rechargeable battery set number of each website is less than capacity less than website of the empty set of cells in this moment and total charge power；

4) the logistics vehicles number that allotment is required every time has vehicle number less than the actual of website；

5) the set of cells sum that each website recalls every time is not more than the full set of cells sum in this moment；

6) the full set of cells flowing between two service stations of single allotment is unidirectional；

7) each allotment, in network total recall set of cells number with total join into set of cells number equal.

Adopt technical scheme, following beneficial effect can be realized: the present invention is directed to the management and running of electric automobile charging and conversion electric network distribution and charging link and carried out basic research, form the basic theories of charging and conversion electric network integration scheduling, operation for charging and conversion electric network provides scientific theory support, give full play to its operational efficiency: (1) sets up concentrated charging station in charging and conversion electric network, the universal model at charging and conversion electric station and dispensing station these three type of service station, introduce in integration scheduling model as a module, when programming without considering the difference in dissimilar service station, simultaneously, universal model has taken into full account the aftereffect that service station state is affected by charging and scheduling batteries behavior, model is made more to become more meticulous；(2) electricity demand correction curve is changed in introducing, from the active service ability in service station, fully take into account the battery replacement device quantity in service station, the average arrival rate of user, acceptable wait to length, change the electricity reality factor such as time exchange power restriction, what make service station changes electricity behavior more realistic meaning；(3) the integrated scheduling model of charging and conversion electric network is set up, the charging scheme of network, battery programs and logistics distribution scheme are carried out global optimization, thus reducing the running cost of network, improve system economy, model is set up based on tou power price simultaneously, the reflection of network load peak valley can effectively be guided again network charging scheme to stabilize network load by tou power price, plays the effect of peak load shifting.

Accompanying drawing explanation

Fig. 1 is the general flow chart of the inventive method；

Fig. 2 is test example transportation network；

Fig. 3 be next day original change electricity requirement forecasting curve；

Fig. 4 is the M/D/C queuing model that electric automobile changes electricity service system；

Fig. 5 is that matlab emulates queuing system operation result；

Fig. 6 be charging and conversion electric station 2 revise before and after change electricity demand curve；

Fig. 7 is that set of cells number asynchronous integration scheduling result is initially expired at charging and conversion electric station；

Fig. 8 is charging and conversion electric station initially empty set of cells number asynchronous integration scheduling result.

Detailed description of the invention

Below technical solution of the present invention is described in detail, but protection scope of the present invention is not limited to described embodiment.

The present embodiment is a kind of electric automobile charging and conversion electric network integration dispatching method based on service station universal model, as it is shown in figure 1, the parameter of embodiment is as follows:

1) transport information is as in figure 2 it is shown, the medium-sized charging station of C presenting set in figure, and G represents charging and conversion electric station.In figure, the numeral of mark is node serial number, is used for characterizing the positional information of website.The corresponding relation of service station numbering and area traffic scattergram is in Table 1, and the dispensing required time of different sections of highway is in Table 2.

Table 1 service station numbering and area traffic scattergram corresponding relation

Table 2 different sections of highway dispensing required time

Starting point	Terminal	Required time/min	Starting point	Terminal	Required time/min	Starting point	Terminal	Required time/min
									1	2	15	7	25	15	15	21	15
1	14	15	8	9	5	16	18	5
									1	16	30	8	17	10	17	21	18
1	20	20	9	10	15	17	23	14
									2	3	15	9	17	10	18	19	13
2	22	15	10	11	10	18	21	15
									3	4	7	10	15	10	19	20	12
3	24	20	10	17	20	19	21	10
									3	25	30	11	12	5	20	22	18
4	5	15	12	13	7	21	22	20
									5	6	5	12	15	12	21	23	16
5	25	10	13	14	10	21	24	25
									6	7	5	13	15	20	22	24	15
6	25	12	13	16	15	23	24	15
									7	8	6	14	16	20	23	25	10
7	17	20	15	16	16
									7	23	25	15	17	15

2) moment every day end, electricity requirement forecasting curve (as shown in Figure 3), the standby full number of batteries of initial time and empty number of batteries (as shown in table 3) transmission are changed to system call center by original for next day in service station.Meanwhile, control centre obtains electricity price information next day from dispatching of power netwoks department, as shown in table 4:

Initial battery pack configuring condition in table 3 network

CF	CE	GF	GE	DF	DE
						0	600	400	600	200	0

Table 4 peak, paddy, flat Time segments division and day part electricity price

3) distribution vehicle load-carrying be 0.9 ton, monoblock battery group quality be 30kg, then a car can load set of cells number is 30 pieces；Each concentrated charging station configuration logistics vehicles 15, each charging and conversion electric station configuration logistics vehicles 5；One complete delivery process includes: 1. load fully charged set of cells at the website that recalls of set of cells, it is necessary to 5min；2. fully charged set of cells is transported to and enter the station a little joining of set of cells；3. joining to enter the station and a little unload full set of cells in set of cells, it is necessary to 5min；4. the set of cells under the replacing of respective numbers is loaded, it is necessary to 5min；5. the set of cells under replacing is transported to dispensing and send website；6. the set of cells under replacing is put on charging rack, it is necessary to 5min；

4) charging process of set of cells is approximately constant output characteristic, and charge power is 2kW, and charging required time is 2.5h；The available charge power of concentrated charging station is 1MW, and charging and conversion electric station can provide charge power to be 0.3MW.The expenses standard of logistics vehicles is 50 yuan/(time *).

(1) service station universal model is set up:

$Q_{i,t}^f=\left\{\begin{array}{cc}{Q}_{i,t-1}^f+Q_{i,t-[t\_\mathrm{chg}/T]}^c-Q_{i,t-1}^h-Q_{i,t-1}^d,& Q_{i,t-1}^d\&GreaterEqual;0\\ Q_{i,t-1}^f+Q_{i,t-[t\_\mathrm{chg}/T]}^c-Q_{i,t-1}^h+\underset{j}{\mathrm{\&Sigma;}}{Q}_{i,t-[{\mathrm{dis}}_{\mathrm{ij}}/T]}^d,& Q_{i,t-[{\mathrm{dis}}_{\mathrm{ij}}/T]}^d<0\end{array}\right.$

$Q_{i,t}^e=\left\{\begin{array}{cc}{Q}_{i,t-1}^e+Q_{i,t-1}^h-Q_{i,t-1}^c+\underset{j}{\mathrm{\&Sigma;}}{Q}_{i,t-[{\mathrm{dis}}_{\mathrm{ij}}*2/T]}^d,& Q_{i,t-[{\mathrm{dis}}_{\mathrm{ij}}*2/T]}^d\&GreaterEqual;0\\ Q_{i,t-1}^e+Q_{i,t-1}^h-Q_{i,t-1}^c-\underset{j}{\mathrm{\&Sigma;}}{Q}_{i,t-[{\mathrm{dis}}_{\mathrm{ij}}/T]}^d,& Q_{i,t-[{\mathrm{dis}}_{\mathrm{ij}}/T]}^d<0\end{array}\right.$

$\left\{\begin{array}{c}{Q}_{i,t}^h\&le;Q_{i,t}^f\\ Q_{i,t}^c\&le;Q_{i,t}^e\\ Q_{i,t}^d\&le;Q_{i,t}^f-Q_{i,t}^h,\mathrm{if}{Q}_{i,t}^d>0\end{array}\right.$

(2) set up electric automobile and change the M/D/C queuing model of electricity service system, as shown in Figure 4.In the present embodiment, there are 6 battery replacement device in service station, and the service time of each car is 4min, Fig. 5 be average arrival rate is 2.2/min, and acceptable congestion lengths is matlab analogue system operation result when 16.It can be seen that user is arrival service station successively within the sampling period, and it is numbered the user of 34,36,43,49,53,54,55,56,57,58,59,60,61,62,63,64,65,66 and fails acquisition and change electricity service.Table 5 list charging and conversion electric station 1 revised change electricity demand curve, Fig. 6 be charging and conversion electric station 2 revise before and after change electricity demand curve.

Table 5 charging and conversion electric station 1 is revised changes electricity curve

Moment	Demand/block	Moment	Demand/block	Moment	Demand/block	Moment	Demand/block
								0:30	6	6:30	1	12:30	12	18:30	7
1:00	3	7:00	0	13:00	5	19:00	12
								1:30	1	7:30	6	13:30	16	19:30	12
2:00	6	8:00	4	14:00	32	20:00	2 6 -->
								2:30	3	8:30	3	14:30	44	20:30	6
3:00	1	9:00	51	15:00	80	21:00	5
								3:30	1	9:30	14	15:30	40	21:30	7
4:00	0	10:00	12	16:00	10	22:00	19
								4:30	1	10:30	7	16:30	8	22:30	4
5:00	2	11:00	38	17:00	3	23:00	11
								5:30	0	11:30	5	17:30	19	23:30	4
6:00	0	12:00	4	18:00	8	0:00	6

(3) according to network traffic figure, utilizing dijkstra algorithm to solve the shortest used time path, table 6 lists the shortest used time path between each service station.

The shortest used time path between each service station of table 6

Initiating station	Target Station	The shortest used time/min	The shortest used time path
				1	6	57	1-2-3-4-5-6
1	11	37	1-14-13-12-11
				1	16	30	1-16
1	17	57	1-14-13-12-15-17
				1	23	58	1-20-19-21-23
1	24	45	1-2-22-24
				6	11	41	6-7-8-9-10-11
6	16	50	6-7-8-17-15-16
				6	17	21	6-7-8-17
6	23	21	6-7-23
				6	24	36	6-7-23-24
11	16	27	11-12-13-16
				11	17	30	11-10-17
11	23	44	11-10-17-23
				11	24	57	11-12-15-21-24
16	17	29	16-15-17

16	23	36	16-18-21-23
				16	24	45	16-18-21-24
17	23	14	17-23
				17	24	29	17-23-24
23	24	15	23-24

(4) logistics cost runing in a few days charging and conversion electric network is calculated:

$F_d=\underset{t}{\mathrm{\&Sigma;}}\underset{i}{\mathrm{\&Sigma;}}\underset{j}{\mathrm{\&Sigma;}}\mathrm{ceil}(\mathrm{abs}\left(Q_{\mathrm{ij},t}^d\right)/\mathrm{vc})*{\mathrm{dis}}_{\mathrm{ij}}*c_{\mathrm{ev}}/120$

In formula: F_dFor logistics cost；The full set of cells number of service station j it is distributed to, if full set of cells is distributed to service station i from service station j, then for t service station iIt is negative；Vc is the set of cells number that each car can load, and unit is block；c_evFor single logistics vehicles freight charges hourly, unit be unit/hour；

(5) a charging expense runing in a few days charging and conversion electric network is calculated:

$F_c=\underset{i}{\mathrm{\&Sigma;}}\underset{\mathrm{time}=1}{\overset{48}{\mathrm{\&Sigma;}}}\underset{t=\mathrm{time}}{\overset{\mathrm{time}+[t\_\mathrm{chg}/30]-1}{\mathrm{\&Sigma;}}}{Q}_{i,t}^c*p_{\mathrm{ack}}*T*p_t$

In formula: F_cFor charging expense；p_ackFor the charge power of set of cells, unit is KW；p_tFor the electricity price of t, unit is unit；

(6) with tou power price for background, with network operation cost minimization for target, actual electric automobile charging and conversion electric network operation integration scheduling model is set up, thus drawing charging and the battery allotment strategy of integration scheduling:

MinF=F_c+F_d

$s.t.Q_{i,t}^h=\min (Q_{i,t}^f,Q_{i,t}^r),\&ForAll;i,t$

$Q_{i,t}^h=Q_{i,t}^r,\&ForAll;i,t$

$Q_{i,t}^c\&le;\min (Q_{i,t}^e,P_i*1000/p_{\mathrm{ack}}),\&ForAll;i,t$

$\underset{j}{\mathrm{\&Sigma;}}\mathrm{ceil}(Q_{\mathrm{ij},t}^d/\mathrm{vc})\&le;{\mathrm{nc}}_i,\mathrm{if}{Q}_{\mathrm{ij},t}^d>0,\&ForAll;i,t$

$Q_{i,t}^d\&le;Q_{i,t}^f-Q_{i,t}^h,\mathrm{if}{Q}_{i,t}^d>0,Q_{i,t}^d=\underset{j}{\mathrm{\&Sigma;}}{Q}_{\mathrm{ij},t}^d,\&ForAll;i,t$

$Q_{\mathrm{ij},t}^d+Q_{\mathrm{ji},t}^d=0$

$\underset{i}{\mathrm{\&Sigma;}}{Q}_{i,t}^d=0,\&ForAll;t$

In formula: P_iFor the capacity of service station i, unit is MW；nc_iLogistics vehicles number for website i.

In the present embodiment, limit charging interval optional set as { 1:00,3:00,8:00,13:00,17:00,21:00}, the optional set of logistics distribution time is { 6:00,9:00,15:30,18:00,22:00}, solve and obtain optimum charging scheme in Table 7, and optimum set of cells programs is in Table 8.

The optimum charging scheme of table 7

The optimum battery programs of table 8

Set out website	Arrive website	Path	Time	Dispensing quantity
					17	6	17-8-7-6	15:30	30 8 -->
1	11	1-14-13-12-11	6:00	90
					17	16	17-15-16	6:00	60
1	24	1-2-22-24	6:00	60
					1	23	1-20-19-21-23	6:00	28
17	23	17-23	6:00	300

The operation total cost of optimal case is 4818.9 yuan, and charging expense is 3858.9 yuan, and logistics cost is 960 yuan.Lacking battery block number is 0, it is possible to meets all of user and changes electricity demand.

Set of cells number asynchronous integration scheduling result such as Fig. 7 is initially expired at charging and conversion electric station, it can be seen that along with the increase of initial completely set of cells number, the general trend of charging expense and systematic running cost is all reduce, logistics cost first reduces and remains unchanged afterwards.Charging and conversion electric station is empty set of cells number asynchronous integration scheduling result such as Fig. 8 initially, it can be seen that along with the increase of initial empty set of cells number, the general trend of charging expense, logistics cost and systematic running cost is all first reduce to remain unchanged afterwards.

As above, although represented and described the present invention with reference to specific preferred embodiment, but it shall not be construed as the restriction to the present invention self.Under the spirit and scope of the present invention premise defined without departing from claims, it can be made in the form and details various change.

Claims (4)

1. the electric automobile charging and conversion electric network integration dispatching method based on service station universal model, it is characterised in that: comprise the following steps:

1) universal model in three types service station in charging and conversion electric network is set up: the universal model of concentrated charging station, charging and conversion electric station and dispensing station, to characterize the relation between state and the behavior in service station, described state includes full set of cells number and empty set of cells number, described behavior includes changing electricity, charging, battery allotment, and the relation between described state and behavior is:

$Q_{i,t}^f=\left\{\begin{array}{cc}{Q}_{i,t-1}^f+Q_{i,t-\&lsqb;t\_chg/T\&rsqb;}^c-Q_{i,t-1}^h-Q_{i,t-1}^d,& Q_{i,t-1}^d\&GreaterEqual;0\\ Q_{i,t-1}^f+Q_{i,t-\&lsqb;t\_chg/T\&rsqb;}^c-Q_{i,t-1}^h+\underset{j}{\mathrm{\&Sigma;}}{Q}_{i,t-\&lsqb;{\mathrm{dis}}_{ij}/T\&rsqb;}^d,& Q_{i,t-\&lsqb;{\mathrm{dis}}_{ij}/T\&rsqb;}^d<0\end{array}\right.---\left(1\right)$

$Q_{i,t}^e=\left\{\begin{array}{cc}{Q}_{i,t-1}^e+Q_{i,t-1}^h-Q_{i,t-1}^c+\underset{j}{\mathrm{\&Sigma;}}{Q}_{i,t-\&lsqb;{\mathrm{dis}}_{ij}*2/T\&rsqb;}^d,& Q_{i,t-\&lsqb;{\mathrm{dis}}_{ij}*2/T\&rsqb;}^d\&GreaterEqual;0\\ Q_{i,t-1}^e+Q_{i,t-1}^h-Q_{i,t-1}^c-\underset{j}{\mathrm{\&Sigma;}}{Q}_{i,t-\&lsqb;{\mathrm{dis}}_{ij}/T\&rsqb;}^d,& Q_{i,t-\&lsqb;{\mathrm{dis}}_{ij}/T\&rsqb;}^d<0\end{array}\right.---\left(2\right)$

$\left\{\begin{array}{c}{Q}_{i,t}^h\&le;Q_{i,t}^f\\ Q_{i,t}^c\&le;Q_{i,t}^e\\ Q_{i,t}^d\&le;Q_{i,t}^f-Q_{i,t}^h,if{Q}_{i,t}^d>0\end{array}\right.---\left(3\right)$

In formula:Full set of cells quantity for t service station i；For the empty set of cells quantity of t service station i, t=1 represents initial time；It is the full set of cells quantity that user changes electricity for t service station i；For the t service station i empty set of cells quantity started to charge up；The full set of cells quantity of allotment is started for t service station i,Represent that t service station i recalls set of cells,Represent that t has full set of cells to recall from other service stations to service station i；T_chg is set of cells charging required time, and unit is min；T is the sampling period, and unit is min；dis_ijSet of cells for service station i to service station j allocates the minimum one way used time, and unit is min；[] expression rounds up；

2) according to the correlation theory of queueing theory, electric automobile changes electricity service system can regard a queuing system as, the client of system is electric automobile, information desk is battery replacement device, the service that information desk provides is to change battery, obtain the actual of service station according to queueing theory principle and change power, the original electricity demand curve that changes is modified:

$Q_{i,t}^r=min(Q_{i,t}^R,N_{i,t})---\left(4\right)$

In formula: N_i,tIn the t sampling period, service station i's changes power；Original for t service station i changes electricity demand；For revise after t service station i change electricity demand；

3) according to network traffic figure, dijkstra algorithm is utilized to solve the shortest used time path；

4) logistics cost runing in a few days charging and conversion electric network is calculated:

$F_d=\underset{t}{\&Sigma;}\underset{i}{\&Sigma;}\underset{j}{\&Sigma;}ceil\left(abs\right(Q_{ij,t}^d)/vc)*{\mathrm{dis}}_{ij}*c_{ev}/120---\left(5\right)$

In formula: F_dFor logistics cost；The full set of cells number of service station j it is distributed to, if full set of cells is distributed to service station i from service station j, then for t service station iIt is negative；Vc is the set of cells number that each car can load, and unit is block；c_evFor single logistics vehicles freight charges hourly, unit be unit/hour；

5) a charging expense runing in a few days charging and conversion electric network is calculated:

$F_c=\underset{i}{\&Sigma;}\underset{time=1}{\overset{48}{\&Sigma;}}\underset{t=time}{\overset{time+\&lsqb;t\_chg/30\&rsqb;-1}{\&Sigma;}}{Q}_{i,t}^c*p_{ack}*T*P\left(t\right)---\left(6\right)$

In formula: F_cFor charging expense；p_ackFor the charge power of set of cells, unit is KW；The electricity price that P (t) is t, unit is unit；

6) with tou power price for background, with network operation cost minimization for target, actual electric automobile charging and conversion electric network operation integration scheduling model is set up, thus drawing charging and the battery allotment strategy of integration scheduling:

MinF=F_c+F_d(7)

$\begin{array}{ccc}s.t.& Q_{i,t}^h=min(Q_{i,t}^f,Q_{i,t}^r)& \&ForAll;i,t\end{array}---\left(8\right)$

$\begin{array}{cc}{Q}_{i,t}^h=Q_{i,t}^r& \&ForAll;i,t\end{array}---\left(9\right)$

$\begin{array}{cc}{Q}_{i,t}^c\&le;min(Q_{i,t}^e,P_i*1000/p_{ack})& \&ForAll;i,t\end{array}---\left(10\right)$

$\begin{array}{ccc}\underset{j}{\&Sigma;}ceil(Q_{ij,t}^d/vc)\&le;{\mathrm{nc}}_i& if{Q}_{ij,t}^d>0& \&ForAll;i,t\end{array}---\left(11\right)$

$\begin{array}{ccc}{Q}_{i,t}^d\&le;Q_{i,t}^f-Q_{i,t}^h& if{Q}_{i,t}^d>0,Q_{i,t}^d=\underset{j}{\&Sigma;}{Q}_{ij,t}^d& \&ForAll;i,t\end{array}---\left(12\right)$

$Q_{ij,t}^d+Q_{ji,t}^d=0---\left(13\right)$

$\begin{array}{cc}\underset{i}{\&Sigma;}{Q}_{i,t}^d=0& \&ForAll;t\end{array}---\left(14\right)$

In formula: P_iFor the capacity of service station i, unit is MW；nc_iLogistics vehicles number for website i.

2. the electric automobile charging and conversion electric network integration dispatching method based on service station universal model according to claim 1, it is characterized in that, step 1) the service station universal model set up discloses the relation of the restriction mutually that influences each other between the state in service station and behavior:

1.1) its state of the behavioral implications in service station change

Instant behavior, immediately the state in service station is produced impact as changed electricity, and process state has aftereffect, namely process state can affect service station state sometime thereafter: 1. charging behavior can affect the empty set of cells number in service station when occurring, but will then through the duration that charges, the full set of cells number in service station just can change；2. when battery allotment starts, only impact recalls the service station of full set of cells, set of cells is packed in the service of reception completely set of cells when arriving of allocating and sky set of cells quantity all changes, and until logistics vehicles returns to the service station recalling full set of cells, its sky set of cells number also changes；

1.2) its behavior of the state constraint in service station

Changing electricity and the battery allotment constraint by full set of cells number, charging is by the constraint of empty set of cells number.

3. the electric automobile charging and conversion electric network integration dispatching method based on service station universal model according to claim 1, it is characterized in that, step 2) in the calculating process of system service ability taken into account system service ability and be subject to the battery replacement device quantity in service station, the average arrival rate of user, acceptable congestion lengths, change the restriction of electricity time factor, specifically include following steps:

2.1) setting up electric automobile and change the M/D/C queuing model of electricity service system, M/D/C represents that quantum condition entropy is obeyed at the client interval time of advent, service time is the model having C information desk in fixed value and system；

2.2) according to service station each sampling period initially change electricity demand, calculate its average arrival rate and the user interval time of advent:

${\mathrm{\&lambda;}}_{i,t}=\frac{Q_{i,t}^R}{T}---\left(15\right)$

$P_{i,t}(T\&le;t)=1-e^{-{\mathrm{\&lambda;}}_{i,t}t}---\left(16\right)$

In formula: λ_i,tIt it is the average arrival rate of service station i in the t sampling period；Be in the t sampling period service station i initially change electricity demand；

2.3) practical operation situation of matlab analogue simulation certain scale system, it is thus achieved that the electric automobile number of service station active service in each sampling period, calculates service rate η:

2.4) calculating the actual power that changes in service station is:

N_i,t=λ_i,t*T*η_i,t(18)。

4. the electric automobile charging and conversion electric network integration dispatching method based on service station universal model according to claim 1, it is characterised in that described step 6) constraints be:

6.1) what service station provided the user change, and electricity changes the smaller value of electricity demand equal to its full set of cells number and user；

6.2) system operation meets and basic changes electricity demand；

6.3) the rechargeable battery set number of each website is less than capacity less than website of the empty set of cells in this moment and total charge power；

6.4) the logistics vehicles number that allotment is required every time has vehicle number less than the actual of website；

6.5) the set of cells sum that each website recalls every time is not more than the full set of cells sum in this moment；

6.6) the full set of cells flowing between two service stations of single allotment is unidirectional；

6.7) each allotment, in network total recall set of cells number with total join into set of cells number equal.