CN106022514A - Public electric bicycle leasing point address-selecting method based on trip chain - Google Patents

Abstract

The invention discloses a public electric bicycle leasing point address-selecting method based on a trip chain. The method comprises steps of: 1) acquiring the planning basic information of a public electric bicycle system to be established; 2) establishing the generalized trip cost function; 3) establishing an upper model and a lower model by considering the planner and the users of the public electric bicycle system respectively so as to construct a double-layer planning model to describe the public electric bicycle leasing point address-selecting problem; and 4) solving the double-layer planning model formed by the upper planning model and the lower planning model to obtain a public electric bicycle leasing point address-selecting scheme. The public electric bicycle leasing point address-selecting method may enhance the accuracy and reliability of the public electric bicycle leasing points, gives an impetus to the construction of associated city public electric bicycle system, and has real popularization value.

Description

A kind of public electronic Cycle Hire point site selecting method based on Trip chain

Technical field

The present invention relates to the public electronic bicycle system planning field in city, refer in particular to a kind of based on Trip chain public Electric bicycle lease point site selecting method.

Background technology

In recent years, public bicycles achieves bigger development in many cities.But cycling trip expends muscle power, rides Apart from limited, in China urban, particularly south tier 2 cities ride experience the best, the most day by day by the most comfortable, laborsaving, Low-carbon (LC), the electric bicycle of distance of riding are replaced.In these cities, develop public electronic bicycle system, beyond doubt Efficent use of resources, meets resident's low cost low-carbon (LC) trip requirements, and the feasible way effectively managed electric bicycle One of.

Lease point addressing is the important step setting up public electronic bicycle system altogether, its most direct decision systems The success or failure built.Different from bicycle, electric bicycle needs charging, and therefore the configuration of charging pile is public electric bicycle system During system addressing it is also contemplated that a key factor, be also simultaneously the key factor affecting system Construction operation cost.Cause This, do not consider that the public bicycles lease point location theory of charging problems is not directly applied for public electronic Cycle Hire point Addressing practice, it is necessary to the site selection model of public electronic Cycle Hire point is studied.

Achievement in research in terms of public bicycles lease point addressing is more the most both at home and abroad, mainly includes experience addressing Method, consideration the hub site site selection model of stock, mixed-integer programming model, center arc site selection model, Bi-level Programming Models etc.. Although public bicycles lease point site selection model is constantly improved and development, but some key issue is the most unresolved: (1) fails to consider " walking, by means of car, ride, return the car, walking " this complete public bicycles uses process；(2) single public bicycles is rented Siting analysis a little of renting is more, fails to be formed the addressing scheme of networking；(3) fail to consider travel time and lease expenses The Trip Costs brought to user；(4) fail to consider the charging problems of electric bicycle.Additionally, there is no at present public affairs both at home and abroad Common-battery moves the research of Cycle Hire point site selection model.

The present invention, from the angle of public electric bicycle network system, introduces the concept of Trip chain, it is considered to common electrical Dynamic interactive effect between bicycle system designer and user, respectively from the angle of system optimal and the equilibrium of user is set up Underlying model, thus obtain a kind of public electronic Cycle Hire point site selecting method based on Trip chain.

Summary of the invention

It is an object of the invention to overcome the shortcoming and defect of prior art, it is provided that a kind of based on Trip chain public electronic Cycle Hire point site selecting method, breaks through traditional failing and considers complete Trip chain, comprehensive travel cost and the list of charging problems Individual public electronic Cycle Hire point site selecting method so that the site selecting method of public electronic Cycle Hire point more rationally section Learning, great impetus is played in the construction to public electronic bicycle system.

For achieving the above object, technical scheme provided by the present invention is: a kind of public electrical salf-walking based on Trip chain Car lease point site selecting method, comprises the following steps:

1) the foundation of planning data setting public electronic bicycle system is planned to build in acquisition, altogether includes the data of five aspects: standby Select public electronic Cycle Hire point data, traffic zone and the range data of lease point, public electric bicycle requirement forecasting Data, construction operation cost data, road network related data；

2) the broad sense Trip Costs function of each Trip chain is set up: first, calculate walking time, residence time respectively and ride Time, build the broad sense Trip Costs function of each Trip chain then in conjunction with decision variable；

3) consider that the interests of public electronic bicycle system designer and user set up upper layer model and underlying model respectively, Thus build Bi-level Programming Models to describe the location problem of public electronic Cycle Hire point；

4) Bi-level Programming Models being made up of upper strata plan model and lower floor's plan model is solved, obtain common electrical Dynamic Cycle Hire point addressing scheme.

In step 1) in, described alternative public electronic Cycle Hire point data include alternative lease point number, position, Service area, electric bicycle turnover rate predictive value, charging pile turnover rate predictive value, user arrive rule etc., from public electronic Bicycle system programme obtains public electronic Cycle Hire point data；Described traffic zone and the range data of lease point Including the distance of the traffic zone centre of form to lease point, the distance etc. between lease point, from public electronic bicycle system planning side Case obtains traffic zone and the distance of lease point；Described public electric bicycle requirement forecasting data refers between each traffic zone Public electronic cycling trip OD amount predictive value (including the predictive value of peak hour), obtains common electrical from traffic volume forecast data Dynamic bicycle requirement forecasting data；Construction operation cost data includes electric bicycle unit price, charging pile unit price, infrastructure construction If expense, operation maintenance expense (including the electricity charge), value hourage, lease expenses etc., designed by market survey and system Scheme can obtain construction operation cost data；Road network related data include walking speed, riding speed, road facilities, Impedance function etc..

In step 2) in, set up the broad sense Trip Costs function of each Trip chain, comprise the steps of:

2.1) Trip chain is set up: Trip chain refers to that user uses public electric bicycle from O point to unidirectional trip of D point Journey, including five stages: 1. walk to by means of car point m from traffic zone r；2. car is being borrowed by means of car point m；3. public electrical salf-walking is used Car borrows car point m to ride to a n that returns the car altogether；4. return the car at a n that returns the car；5. traffic zone s is walked to from a n that returns the car.According to this Definition, sets up all of Trip chain in public electric bicycle network system.

2.2) calculate the walking time: the walking time be user from O point to by means of car point or from return the car a little to used by D point time Between.Assuming that traffic zone trip requirements is uniformly distributed, therefore for whole system, trip can be regarded as from the centre of form to another The process of the traffic zone centre of form.Therefore, from traffic zone r to the walking time by means of car point m it is:

$t_{rm}=\frac{S_{rm}}{walking\_speed}$

In formula: S_rmFor the centre of form of traffic zone r to the distance by means of car point m；Walking_speed is walking speed.

Walking time t from the r to traffic zone s that returns the car_nsIn like manner can calculate.

2.3) residence time is calculated: the scale of lease point is true by the public electric bicycle number of charging pile number and outfit Fixed, both computing formula are as follows:

In formula: u_mFor charging pile quantity；b_mFor the public electric bicycle quantity being equipped with；q_mCommon electrical for the m by means of car point Dynamic bicycle flow, Charging pile turnover rate for lease point m；For lease point m Bicycle turnover rate.

Residence time is that user is leasing the waiting time of point and time sum of borrow/returning the car.Public electronic Cycle Hire Point can regard as power system capacity conditional multichannel queuing system, is assuming that client obeys Poisson flow, during the service of each charging pile Between obey quantum condition entropy and work separate under conditions of, the computing formula of residence time of lease point m is:

$t_m=\frac{{{\mathrm{\ρ}}_m}^{u_m}{\mathrm{\β}}_m{p}_{0m}}{q_m{u}_m!{(1-{\mathrm{\β}}_m)}^2}+\frac{1}{{\mathrm{\μ}}_m}$

In formula: μ_m: the service frequency of system, its computing formula isρ_m: the service intensity of system, meter Calculation formula is ρ_m=q_m/(24*μ_m)；β_m: the facility utilization rate of system, computing formula is β=ρ_m/u_m；K: accepting the client of service Quantity.p_0m: the unmanned probability accepting service in system, computing formula is

Return the car t residence time of a n_nIn like manner can calculate.

2.4) calculate and ride the time: the timing definition of riding of user be user by means of car point and between returning the car a little when riding Between.Owing to the running time of bicycle is affected less by bicycle flow, and affected relatively big by stroke distances, therefore It is believed that the impedance approximately equal in each section in same Trip chain, the impedance difference of different Trip chain is mainly by the difference of its distance Cause.Therefore, its computing formula of time of riding is as follows.

In formula: t₀The running time freely flowed for electric motor car；C_kThe electronic wagon flow of maximum passed through for certain energy ridden on section Amount；α, β are parameter to be calibrated；S_mnSpacing for lease point m, n；ξ is electrical salf-walking under the conditions of Computer method or the immiscible row of people The speed reduction coefficient of car；V is travel speed；q_mnFor the public electronic bicycle flow amount between website m, n, computing formula is

2.5) calculate the broad sense Trip Costs function of each Trip chain: user OD to (r, s) between the trip of kth bar Trip chain Broad sense Trip Costs be:

$t_k\left(q_k^{rs}\right)=(y_m{t}_{rm}+y_m{\mathrm{\δ}}_{m,k}^{rs}{t}_m+y_m{y}_n{t}_{mn}+y_n{\mathrm{\δ}}_{n,k}^{rs}{t}_n+y_n{t}_{ns})+c/\mathrm{\τ}$

Wherein,OD to (r, s) between kth bar Trip chain on public electronic bicycle flow amount；t_rm: user is from traffic Community r is to the walking time by means of car point m；t_m: user in the residence time by means of car point m, m ∈ K；t_mn: user from by means of car point m to also Car point n rides the time；t_n: user in the residence time returning the car a n, n ∈ K；t_nsFor user from the n to traffic zone s that returns the car Walking time；τ is to be worth hourage；C is lease expenses；K is that alternative public bicycles lease point (includes by means of car point and goes back Car point) set；For Trip chain lease point relation variable, if lease point m OD to (r, s) in kth bar Trip chain,Otherwise, In like manner.y_mFor decision variable, lease some m, y if building_m=1, otherwise, y_m=0；y_nWith Reason.If the relation of decision variable and Trip chain lease point relation variable is y_a=0, then

In step 3) in, set up Bi-level Programming Models to describe the location problem of public electronic Cycle Hire point, comprise Following steps:

3.1) layer model in foundation: upper layer model is system optimal problem, the constraints that needs meet is: 1. two lease Distance between point must be at [d_min,d_maxIn the range of]；The most at least select one and borrow car point；The most at least select one to return the car a little； 4. lease point construction fund sum is not more than the total investment upper limit；Public electronic cycling trip amount sum in the most each Trip chain Measure equal to electric bicycle trip requirements OD between community；Public electronic cycling trip amount in the most each Trip chain is non-negative Value.

U:

Subject to:

d_min≤y_my_nd_mn≤d_max(m≠n)

Σ_m∈Ky_m≥1

∑_n∈Ky_n≥1

∑_m∈Ky_mi_m+∑_n∈Ky_ni_n≤1

$\begin{array}{cc}{\mathrm{\Σ}}_{k\∈K}{q}_k^{rs}=q_{rs}& \∀r,s\end{array}$

$\begin{array}{cc}{q}_k^{rs}\≥0& \∀r,s\end{array}$

In formula:

d_minAcceptable minimum interval for lease point；d_maxFor the maximum allowable spacing of lease point, take electric bicycle once Maximum range after charging；I is the total investment upper limit；i_m,i_nIt is respectively each construction operation by means of car point m with a n that returns the car Expense.

3.2) setting up underlying model: underlying model is user equilibrium model, the condition that needs meet is: in the most each Trip chain Public electronic cycling trip amount sum equal to the electric bicycle trip requirements OD amount between community；In the most each Trip chain Public electronic cycling trip amount is nonnegative value.

L:

Subject to:

$\begin{array}{cc}{\mathrm{\Σ}}_{k\∈K}{q}_k^{rs}=q_{rs}& \∀r,s\end{array}$

$\begin{array}{cc}{q}_k^{rs}\≥0& \∀r,s\end{array}$

In formula: q_rsFor OD to (r, s) between public e-bike traffic amount.

In step 4) in, the Bi-level Programming Models being made up of upper strata plan model and lower floor's plan model is solved, Obtain public electronic Cycle Hire point addressing scheme, comprise the steps of:

4.1) iterations z=0 is made；Assume that lease site is the most selected,Each Trip chain public Electrical salf-walking vehicle flowrate is 0,

4.2) by upper strata solution to modelSubstitute in underlying model, use F-W Algorithm for Solving underlying model, obtain The solution of underlying model

4.3) by the solution of underlying modelIn substitution in layer model, use Hybrid Particle Swarm Optimization Solve layer model, obtain upper strata solution to model

4.4) whether judged result meets iteration stopping condition, if meeting, then exports globally optimal solution；If being unsatisfactory for, then Return step 4.2) continue to solve；

4.5) according to model solution result, obtain public electronic Cycle Hire point addressing scheme and choose filling of lease point Electricity stake number and the public electric bicycle number of outfit.

The present invention compared with prior art, has the advantage that and beneficial effect:

The present invention, from the angle of public electric bicycle network system, introduces the concept of Trip chain, it is considered to common electrical Dynamic interactive effect between bicycle system designer and user, respectively from the angle of system optimal and the equilibrium of user is set up Underlying model, thus obtain a kind of public electronic Cycle Hire point site selecting method based on Trip chain.First to public electronic The lease point site selection model of bicycle system is studied；It is firstly introduced the concept of Trip chain, sets up Bi-level Programming Models, it is achieved The balance of interest of system planner and user；Consider the travel time and Trip Costs that lease expenses brings；Can root According to the flexible performance model of the situation of various places, strong adaptability, there is stronger promotional value.

Accompanying drawing explanation

Fig. 1 is the workflow diagram of the present invention.

Fig. 2 is alternative public electric bicycle system lease point schematic diagram in embodiment.

Detailed description of the invention

Below in conjunction with specific embodiment, the invention will be further described.

As it is shown in figure 1, the public electronic Cycle Hire point site selecting method described in the present embodiment, comprise the following steps:

1) the foundation of planning data setting public electronic bicycle system is planned to build in acquisition, altogether includes the data of five aspects: standby Select public electronic Cycle Hire point data, traffic zone and the range data of lease point, public electric bicycle requirement forecasting Data, construction operation cost data, road network related data.Wherein, described alternative public electronic Cycle Hire point data includes The number of alternative lease point, position, service area, electric bicycle turnover rate predictive value, charging pile turnover rate predictive value, user Arrive rule etc., obtain public electronic Cycle Hire point data from public electronic bicycle system programme；Described traffic The range data of community and lease point includes the traffic zone centre of form distance to lease point, and the distance etc. between lease point, from public affairs Common-battery moves bicycle system programme and obtains traffic zone and the distance of lease point；Described public electric bicycle requirement forecasting Data refers to public electronic cycling trip OD amount predictive value (including the predictive value of peak hour) between each traffic zone, from friendship Flux prediction data obtains public electric bicycle requirement forecasting data；Construction operation cost data includes electric bicycle list Valency, charging pile unit price, infrastructure construction expense, operation maintenance expense (including the electricity charge), value hourage, lease expenses Deng, construction operation cost data can be obtained by market survey and system design scheme；Road network related data includes walking speed Degree, riding speed, road facilities, impedance function etc..The data collected is as follows:

Certain university plan set up public electric bicycle rental system, planning region is divided into two traffic zones (M and N).The service area of alternative public electronic Cycle Hire point in each traffic zone and each lease point is as shown in Figure 2.Two Potential public electric bicycle OD amount between individual traffic zone is (wherein the peak hour is 200) on the 2000/th, and range information is such as Shown in table, table, it is as shown in the table for cost information, and it is as shown in the table for other relevant parameters.

The table 1 community centre of form is to the distance (unit: m) of lease point

M1

M2

M3

M4

N1

N2

N3

N4

N5

M

300

210

505

340

/

/

/

/

/

N

/

/

/

/

400

350

60

350

400

Table 2 lease point is to the distance (unit: m) of lease point

	N1	N2	N3	N4	N5
						M1	4110	3513	3600	2178	3613
M2	3916	3100	1869	1976	3650
						M3	3780	4293	4406	3004	4230
M4	4200	3508	2980	3256	4130

Table 3 lease point cost information

Project	Price
		Electric motor car unit price cb	1800 yuan/
Intelligent charging spot unit price cp	4000 yuan/
		Infrastructure construction ci	500 yuan/day
Running cost (including the electricity charge) co	4200 yuan// year
		Average annual operation cost i of one lease point	I=cb·bm+cp·um+365ci+co

Remaining relevant input parameter of table 4

2) set up the broad sense Trip Costs function of each Trip chain, comprise the steps of:

2.1) Trip chain is set up: Trip chain refers to that user uses public electric bicycle from O point to unidirectional trip of D point Journey, including five stages: 1. walk to by means of car point m from traffic zone r；2. car is being borrowed by means of car point m；3. public electrical salf-walking is used Car borrows car point m to ride to a n that returns the car altogether；4. return the car at a n that returns the car；5. traffic zone s is walked to from a n that returns the car.According to this Definition, sets up all of Trip chain in public electric bicycle network system.

2.2) calculate the walking time: the walking time be user from O point to by means of car point or from return the car a little to used by D point time Between.Assuming that traffic zone trip requirements is uniformly distributed, therefore for whole system, trip can be regarded as from the centre of form to another The process of the traffic zone centre of form.Therefore, from traffic zone r to the walking time by means of car point m it is:

$t_{rm}=\frac{S_{rm}}{walking\_speed}$

In formula: S_rmFor the centre of form of traffic zone r to the distance by means of car point m；Walking_speed is walking speed.

Walking time t from the r to traffic zone s that returns the car_nsIn like manner can calculate.

2.3) residence time is calculated: the scale of lease point is true by the public electric bicycle number of charging pile number and outfit Fixed, both computing formula are as follows:

In formula: u_mFor charging pile quantity；b_mFor the public electric bicycle quantity being equipped with；q_mCommon electrical for the m by means of car point Dynamic bicycle flow, Charging pile turnover rate for lease point m；For lease point m Bicycle turnover rate.

Residence time is that user is leasing the waiting time of point and time sum of borrow/returning the car.Public electronic Cycle Hire Point can regard as power system capacity conditional multichannel queuing system, is assuming that client obeys Poisson flow, during the service of each charging pile Between obey quantum condition entropy and work separate under conditions of, the computing formula of residence time of lease point m is:

$t_m=\frac{{{\mathrm{\ρ}}_m}^{u_m}{\mathrm{\β}}_m{p}_{0m}}{q_m{u}_m!{(1-{\mathrm{\β}}_m)}^2}+\frac{1}{{\mathrm{\μ}}_m}$

In formula: μ_m: the service frequency of system, its computing formula isρ_m: the service intensity of system, meter Calculation formula is ρ_m=q_m/(24*μ_m)；β_m: the facility utilization rate of system, computing formula is β=ρ_m/u_m；K: accepting the client of service Quantity.p_0m: the unmanned probability accepting service in system, computing formula is

Return the car t residence time of a n_nIn like manner can calculate.

2.4) calculate and ride the time: the timing definition of riding of user be user by means of car point and between returning the car a little when riding Between.Owing to the running time of bicycle is affected less by bicycle flow, and affected relatively big by stroke distances, therefore It is believed that the impedance approximately equal in each section in same Trip chain, the impedance difference of different Trip chain is mainly by the difference of its distance Cause.Therefore, its computing formula of time of riding is as follows.

In formula: t₀The running time freely flowed for electric motor car；C_kThe electronic wagon flow of maximum passed through for certain energy ridden on section Amount；α, β are parameter to be calibrated；S_mnSpacing for lease point m, n；ξ is electrical salf-walking under the conditions of Computer method or the immiscible row of people The speed reduction coefficient of car；V is travel speed；q_mnFor the public electronic bicycle flow amount between website m, n, computing formula is

2.5) calculate the broad sense Trip Costs function of each Trip chain: user OD to (r, s) between the trip of kth bar Trip chain Broad sense Trip Costs be:

$t_k\left(q_k^{rs}\right)=(y_m{t}_{rm}+y_m{\mathrm{\δ}}_{m,k}^{rs}{t}_m+y_m{y}_n{t}_{mn}+y_n{\mathrm{\δ}}_{n,k}^{rs}{t}_n+y_n{t}_{ns})+c/\mathrm{\τ}$

Wherein,OD to (r, s) between kth bar Trip chain on public electronic bicycle flow amount；t_rm: user is from traffic Community r is to the walking time by means of car point m；t_m: user in the residence time by means of car point m, m ∈ K；t_mn: user from by means of car point m to also Car point n rides the time；t_n: user in the residence time returning the car a n, n ∈ K；t_nsFor user from the n to traffic zone s that returns the car Walking time；τ is to be worth hourage；C is lease expenses；K is that alternative public bicycles lease point (includes by means of car point and goes back Car point) set；For Trip chain lease point relation variable, if lease point m OD to (r, s) in kth bar Trip chain,Otherwise, In like manner.y_mFor decision variable, lease some m, y if building_m=1, otherwise, y_m=0；y_nWith Reason.If the relation of decision variable and Trip chain lease point relation variable is y_a=0, then

3) set up Bi-level Programming Models to describe the location problem of public electronic Cycle Hire point, comprise the steps of:

3.1) layer model in foundation: upper layer model is system optimal problem, the constraints that needs meet is: 1. two lease Distance between point must be at [d_min,d_maxIn the range of]；The most at least select one and borrow car point；The most at least select one to return the car a little； 4. lease point construction fund sum is not more than the total investment upper limit；Public electronic cycling trip amount sum in the most each Trip chain Measure equal to electric bicycle trip requirements OD between community；Public electronic cycling trip amount in the most each Trip chain is non-negative Value.

U:

Subject to:

d_min≤y_my_nd_mn≤d_max(m≠n)

∑_m∈Ky_m≥1

∑_n∈Ky_n≥1

∑_m∈Ky_mi_m+∑_n∈Ky_ni_n≤1

$\begin{array}{cc}{\mathrm{\Σ}}_{k\∈K}{q}_k^{rs}=q_{rs}& \∀r,s\end{array}$

$\begin{array}{cc}{q}_k^{rs}\≥0& \∀r,s\end{array}$

In formula:

d_minAcceptable minimum interval for lease point；d_maxFor the maximum allowable spacing of lease point, take electric bicycle once Maximum range after charging；I is the total investment upper limit；i_m,i_nIt is respectively each construction operation by means of car point m with a n that returns the car Expense.

3.2) setting up underlying model: underlying model is user equilibrium model, the condition that needs meet is: in the most each Trip chain Public electronic cycling trip amount sum equal to the electric bicycle trip requirements OD amount between community；In the most each Trip chain Public electronic cycling trip amount is nonnegative value.

L:

Subject to:

$\begin{array}{cc}{\mathrm{\Σ}}_{k\∈K}{q}_k^{rs}=q_{rs}& \∀r,s\end{array}$

$\begin{array}{cc}{q}_k^{rs}\≥0& \∀r,s\end{array}$

In formula: q_rsFor OD to (r, s) between public e-bike traffic amount.

In step 4) in, the Bi-level Programming Models being made up of upper strata plan model and lower floor's plan model is solved, Obtain public electronic Cycle Hire point addressing scheme, comprise the steps of:

4.1) iterations z=0 is made；Assume that lease site is the most selected,Each Trip chain public Electrical salf-walking vehicle flowrate is 0,

4.2) by upper strata solution to modelSubstitute in underlying model, use F-W Algorithm for Solving underlying model, obtain The solution of underlying model

4.3) by the solution of underlying modelIn substitution in layer model, use Hybrid Particle Swarm Optimization Solve layer model, obtain upper strata solution to model

4.4) whether judged result meets iteration stopping condition, if meeting, then exports globally optimal solution；If being unsatisfactory for, then Return step 4.2) continue to solve；

4.5) according to model solution result, obtain public electronic Cycle Hire point addressing scheme and choose filling of lease point Electricity stake number and the public electric bicycle number of outfit.

Algorithm uses matlab language to realize.The penalty function structure rule of upper layer model is as follows: if the constraints of not meeting 1., 2. or 3., penalizing 10000, if not meeting constraints 4., penalizing 30000.The population of Hybrid Particle Swarm Optimization takes 20 Individual, maximum iteration time takes 10000 times.By in above-mentioned basic data input program, what algorithm ran the results are shown in Table.

Table 5 algorithm operation result

Alternative website	Decision variable	Charging pile number	Electric bicycle number
				M1	1	13	18
M2	1	19	26
				M3	0	0	0
M4	0	0	0
				N1	0	0	0
N2	0	0	0
				N3	0	0	0
N4	1	26	37
				N5	0	0	0

From solving result it will be seen that M1 (13 charging piles and 18 public electrical salf-walkings should be built at traffic zone M Car), M2 (19 charging piles and 26 public electric bicycles) two lease point, (26 are filled should to set up N4 at traffic zone N Electricity stake and 37 public electric bicycles) lease point.M1 is subway station, uses public electric bicycle demand of plugging into bigger.M2 For school administration building and afforestation square, for transport need aggregate site.N4, near the centre of form of traffic zone M, has concentrated student herein The living facilities such as dormitory, students' dining hall, sports ground, the transport need of Shi Gai community assembles ground.Therefore, model described herein is used Lease point addressing result be closer to practical situation, there is certain operability and realistic meaning.

In sum, after using above scheme, the present invention is that the addressing of public electronic Cycle Hire point provides newly Method, it is possible to be effectively improved the accuracy and reliability of public electronic Cycle Hire point, effectively promote relevant city public The building-up work of electric bicycle system, has actual promotional value, is worthy to be popularized.

Embodiment described above is only the preferred embodiments of the invention, not limits the practical range of the present invention with this, therefore The change that all shapes according to the present invention, principle are made, all should contain within the scope of the present invention.

Claims (5)

1. a public electronic Cycle Hire point site selecting method based on Trip chain, it is characterised in that comprise the following steps:

1) obtain and plan to build the foundation of planning data setting public electronic bicycle system, altogether include the data of five aspects: alternative public affairs Common-battery move Cycle Hire point data, traffic zone and the range data of lease point, public electric bicycle requirement forecasting data, Construction operation cost data, road network related data；

2) set up the broad sense Trip Costs function of each Trip chain: first, calculate walking time, residence time and the time of riding respectively Between, the broad sense Trip Costs function of each Trip chain is built then in conjunction with decision variable；

3) consider that the interests of public electronic bicycle system designer and user set up upper layer model and underlying model respectively, thus Build Bi-level Programming Models to describe the location problem of public electronic Cycle Hire point；

4) Bi-level Programming Models being made up of upper strata plan model and lower floor's plan model is solved, obtain public electronic from Driving lease point addressing scheme.

A kind of public electronic Cycle Hire point site selecting method based on Trip chain the most according to claim 1, its feature It is: in step 1) in, described alternative public electronic Cycle Hire point data includes the number of alternative lease point, position, clothes Business scope, electric bicycle turnover rate predictive value, charging pile turnover rate predictive value, user arrive rule, from public electrical salf-walking Car systems organization scheme obtains public electronic Cycle Hire point data；The range data of described traffic zone and lease point includes The traffic zone centre of form, to the distance between the distance and lease point of lease point, obtains from public electronic bicycle system programme Traffic zone and the distance of lease point；Described public electric bicycle requirement forecasting data refers to the common electrical between each traffic zone Dynamic cycling trip OD measures predictive value, obtains public electric bicycle requirement forecasting data from traffic volume forecast data；Described build If operation cost data includes electric bicycle unit price, charging pile unit price, infrastructure construction expense, operation maintenance expense, trip The row time value, lease expenses, be obtained in that construction operation cost data by market survey and system design scheme；Described road Net related data includes walking speed, riding speed, road facilities, impedance function.

A kind of public electronic Cycle Hire point site selecting method based on Trip chain the most according to claim 1, its feature It is, in step 2) in, set up the broad sense Trip Costs function of each Trip chain, comprise the steps of:

2.1) set up Trip chain: Trip chain refer to user use public electric bicycle from the unidirectional trip process of O point to D point, Including five stages: 1. walk to by means of car point m from traffic zone r；2. car is being borrowed by means of car point m；3. public electric bicycle is used Car point m is borrowed to ride to a n that returns the car altogether；4. return the car at a n that returns the car；5. traffic zone s is walked to from a n that returns the car；Fixed according to this Justice, sets up all of Trip chain in public electric bicycle network system；

2.2) calculate the walking time: the walking time be user from O point to by means of car point or from returning the car a little to the time used by D point；False If traffic zone trip requirements is uniformly distributed, therefore for whole system, trip can be regarded as and hand over to another from the centre of form The process of the logical community centre of form；Therefore, from traffic zone r to the walking time by means of car point m it is:

$t_{rm}=\frac{S_{rm}}{walking\_speed}$

In formula: S_rmFor the centre of form of traffic zone r to the distance by means of car point m；Walking_speed is walking speed；

Walking time t from the r to traffic zone s that returns the car_nsIn like manner can calculate；

2.3) residence time is calculated: the scale of lease point is determined by the public electric bicycle number of charging pile number and outfit, Both computing formula are as follows:

In formula: u_mFor charging pile quantity；b_mFor the public electric bicycle quantity being equipped with；q_mFor the m by means of car point public electronic from Driving flow, Charging pile turnover rate for lease point m；Oneself of m is put for lease Driving turnover rate；

Residence time is that user is leasing the waiting time of point and time sum of borrow/returning the car；Public electronic Cycle Hire point regards Making power system capacity conditional multichannel queuing system, assuming that client obeys Poisson flow, each charging pile is obeyed negative service time Exponential and work separate under conditions of, the computing formula of residence time of lease point m is:

$t_m=\frac{{{\mathrm{\ρ}}_m}^{u_m}{\mathrm{\β}}_m{p}_{0m}}{q_m{u}_m!{(1-{\mathrm{\β}}_m)}^2}+\frac{1}{{\mathrm{\μ}}_m}$

In formula: μ_mFor the service frequency of system, its computing formula isρ_mFor the service intensity of system, calculate Formula is ρ_m=q_m/(24*μ_m)；β_mFor the facility utilization rate of system, computing formula is β=ρ_m/u_m；K is the client accepting service Quantity；p_0mFor the probability accepting service unmanned in system, computing formula is

Return the car t residence time of a n_nIn like manner can calculate；

2.4) calculate and ride the time: the timing definition of riding of user is user by means of car point and riding the time between returning the car a little； Its computing formula of time of riding is as follows:

In formula: t₀The running time freely flowed for electric motor car；C_kThe maximum electric motor car flow passed through for certain energy ridden on section； α, β are parameter to be calibrated；S_mnSpacing for lease point m, n；ξ is electric bicycle under the conditions of Computer method or the immiscible row of people Speed reduction coefficient；V is travel speed；q_mnFor the public electronic bicycle flow amount between website m, n, computing formula is

2.5) calculate the broad sense Trip Costs function of each Trip chain: user OD to (r, s) between kth bar Trip chain trip wide Justice Trip Costs is:

$t_k\left(q_k^{\mathrm{rs}}\right)=(y_m{t}_{\mathrm{rm}}+y_m{\mathrm{\δ}}_{m,k}^{\mathrm{rs}}{t}_m+y_m{y}_n{t}_{\mathrm{mn}}+y_n{\mathrm{\δ}}_{n,k}^{\mathrm{rs}}+y_n{t}_{\mathrm{ns}})+c/\mathrm{\τ}$

Wherein,For OD to (r, s) between kth bar Trip chain on public electronic bicycle flow amount；t_rmLittle from traffic for user District r is to the walking time by means of car point m；t_mFor user in the residence time by means of car point m, m ∈ K；t_mnFor user from by means of car point m to also Car point n rides the time；t_nFor user in the residence time returning the car a n, n ∈ K；t_nsFor user from the n to traffic zone s that returns the car Walking time；τ is to be worth hourage；C is lease expenses；K is the set of alternative public bicycles lease point；For going out Row chain lease point relation variable, if lease point m OD to (r, s) in kth bar Trip chain,Otherwise, In like manner；y_mFor decision variable, lease some m, y if building_m=1, otherwise, y_m=0；y_nIn like manner；Decision variable and Trip chain If the relation of lease point relation variable is y_a=0, then

A kind of public electronic Cycle Hire point site selecting method based on Trip chain the most according to claim 1, its feature It is, in step 3) in, set up Bi-level Programming Models to describe the location problem of public electronic Cycle Hire point, comprise as follows Step:

3.1) layer model in foundation: upper layer model is system optimal problem, the constraints that needs meet is: 1. 2 lease point Between distance must be at [d_min,d_maxIn the range of]；The most at least select one and borrow car point；The most at least select one to return the car a little；4. rent A construction fund sum of renting is not more than the total investment upper limit；Public electronic cycling trip amount sum in the most each Trip chain is equal to Electric bicycle trip requirements OD amount between community；Public electronic cycling trip amount in the most each Trip chain is nonnegative value；

U:

Subject to:

d_min≤y_my_nd_mn≤d_max(m≠n)

∑_m∈Ky_m≥1

∑_n∈Ky_n≥1

∑_m∈Ky_mi_m+∑_n∈Ky_ni_n≤1

$\begin{array}{cc}{\mathrm{\Σ}}_{k\∈K}{q}_k^{rs}=q_{rs}& \∀r,s\end{array}$

$\begin{array}{cc}{q}_k^{rs}\≥0& \∀r,s\end{array}$

In formula:

d_minAcceptable minimum interval for lease point；d_maxFor the maximum allowable spacing of lease point, take electric bicycle and once charge After maximum range；I is the total investment upper limit；i_m,i_nIt is respectively each construction operation expense by means of car point m with a n that returns the car With；

3.2) setting up underlying model: underlying model is user equilibrium model, the condition that needs meet is: the public affairs in the most each Trip chain Common-battery moves cycling trip amount sum equal to the electric bicycle trip requirements OD amount between community；Public in the most each Trip chain Electric bicycle travel amount is nonnegative value；

L:

Subject to:

$\begin{array}{cc}{\mathrm{\Σ}}_{k\∈K}{q}_k^{rs}=q_{rs}& \∀r,s\end{array}$

$\begin{array}{cc}{q}_k^{rs}\≥0& \∀r,s\end{array}$

In formula: q_rsFor OD to (r, s) between public e-bike traffic amount.

A kind of public electronic Cycle Hire point site selecting method based on Trip chain the most according to claim 1, its feature It is, in step 4) in, the Bi-level Programming Models being made up of upper strata plan model and lower floor's plan model is solved, obtains Public electronic Cycle Hire point addressing scheme, comprises the steps of:

4.1) iterations z=0 is made；Assume that lease site is the most selected,Each Trip chain public electronic from Driving flow is 0,

4.2) by upper strata solution to modelSubstitute in underlying model, use F-W Algorithm for Solving underlying model, obtain lower floor Solution to model

4.3) by the solution of underlying modelIn substitution in layer model, Hybrid Particle Swarm Optimization is used to solve Upper layer model, obtains upper strata solution to model

4.4) whether judged result meets iteration stopping condition, if meeting, then exports globally optimal solution；If being unsatisfactory for, then return Step 4.2) continue to solve；

4.5) according to model solution result, obtain public electronic Cycle Hire point addressing scheme and choose the charging pile of lease point Number and the public electric bicycle number of outfit.