CN111651899A - Robust location and capacity determination method and system for battery swap stations considering user selection behavior - Google Patents

Abstract

The invention discloses a method and system for robust location selection and capacity determination of a power exchange station considering user selection behavior, wherein the method includes: establishing a user selection behavior model based on multiple Logit models, constructing a location selection model, and equivalently transforming to form SOCP constraints; The distributed robust optimization method is used to deal with uncertain parameters, expressing the minimum number of batteries that meet the preset service level, and equivalently transforming to form SOCP constraints; expressing the minimum number of battery-swapping robots that meet the preset service level, and equivalently transforming to form SOCP constraints; The objective function of SOCP constraint and revenue maximization is based on the MISOCP model to build a distributed robust location and capacity model for battery swapping stations. The solver is called to solve the problem, and the results of the location and capacity of battery swapping stations are obtained. Through the technical scheme of the present invention, the user's selection behavior and uncertainty are comprehensively considered, and profits are maximized at the same time, which improves the scientificity of site selection, station construction and volume determination.

Description

Robust location and capacity determination method and system for battery swap stations considering user selection behavior

technical field

The invention relates to the technical field of operation management and operations research, in particular to a method for robust location and capacity determination of a power exchange station considering user selection behavior and a robust site selection and capacity determination system for power exchange substations considering user selection behavior.

Background technique

As the global energy crisis and climate change intensify, countries around the world are looking for solutions. In the field of transportation, the promotion of electric vehicles has received strong support. Currently, there are two modes of electric vehicle energy supplementation: charging mode and battery swapping mode. However, the charging mode has the following disadvantages: long charging time, high investment cost and limited vehicle mileage. In order to solve the shortage of electric vehicle charging mode, the electric vehicle industry began to explore the battery swap mode. Although the battery swap model failed in 2013 due to the bankruptcy of Better Place. But in recent years, many companies have begun to re-explore the battery swap model. For example, BAIC BJEV launched the "Optimus Plan", which aims to deeply integrate new energy vehicles, power batteries, power swap stations, and photovoltaic power generation. , put 500,000 battery swap vehicles. Since 2019, BAIC and Aodong have established a strategic cooperation direction of "deepening integration vertically and rapid development horizontally" at the group level, and jointly promote the comprehensive promotion and application of the battery swap model. In the next five years, Aodong plans to enter 50 cities across the country, build 5,000 battery swap stations, support the battery swap operation of 2 million new energy vehicles, and provide cities with more efficient energy supply services.

The location and capacity of electric vehicle swapping stations are crucial to the operation of battery swapping services. Regarding site selection, the earliest deterministic site selection model appeared. Deterministic location models can be divided into three categories: node-based location models, arc-based location models, and path-based location models. Currently, most siting models are path-based models. Hodgson first began to study the path-based location model. He proposed the Flow Capturing Location (FCL) model. Under the constraints of the given number of stations, the goal of the model is to maximize the given starting and ending points (Origin -Destination, OD) total demand for inter-service. The model assumes that as long as there is one stop on a route, the demand on this route can be serviced, but the model does not take into account the limited range of the vehicle. After Hodgson proposed the FCL model, many scholars proposed improved models: the Flow Refueling Location (FRL) model considering the vehicle cruising range, the Constrained Flow Refueling Location (CFRL) considering the station capacity constraints ) model, etc. However, the above models all need to pre-generate a large number of feasible site selection point combinations, so some scholars have improved them and proposed new models: new node-based and arc-based FRL models, and extended network models. But none of these models take into account demand uncertainty, user choice behavior, and service level requirements.

SUMMARY OF THE INVENTION

In view of the above problems, the present invention provides a method and system for robust site selection and capacity determination of a power exchange station considering user selection behavior. The minimum number of batteries and the minimum number of battery swapping robots required, under the objective function of maximizing revenue, build a robust distribution model of location and capacity for battery swapping stations, and equivalently convert it into a mixed integer second-order cone programming (Mixed Integer Second Order ConeProgramming, MISOCP) model, so as to obtain the results of location selection and capacity determination of electric vehicle swap station. The method provided by the invention comprehensively considers uncertainty and user selection behavior, and at the same time, it can maximize the profits of operators in site selection and construction, battery and battery swap robots, and improve the scientificity of site selection and construction and capacity determination.

In order to achieve the above object, the present invention provides a robust method for site selection and capacity determination of a power exchange station considering user selection behavior, including: establishing a user selection behavior model based on multiple Logit models for the behavior of the user selecting a site to obtain power exchange services; Based on the user selection behavior model, a site selection model is constructed; the site selection model is equivalently transformed into a first second order cone programming (Second Order Cone Programming, SOCP) constraint; a distributed robust optimization method is used to process uncertain parameters, And express the minimum number of batteries required by the site to satisfy the preset service level; convert the expression of the required minimum number of batteries into a second SOCP constraint equivalently; according to the user selection decision under the user selection behavior model, express the Set the minimum number of battery-changing robots required by the site under the service level; equivalently transform the expression of the required minimum number of battery-changing robots to form a third SOCP constraint; combine the first SOCP constraint, the second SOCP constraint, and all The third SOCP constraint and the objective function of maximizing revenue are described, and the MISOCP model is used to build a distributed robust location and capacity model for battery swap stations; the solver is called to solve the distributed robust location and capacity model for battery swap stations, and the The results of site selection and capacity determination of electric vehicle swap stations.

In the above technical solution, preferably, the preset service level satisfied by the expression of the minimum number of batteries required by the site is: the probability that the state of charge (State of Charge, SOC) of the swapped battery is not lower than the preset value is not less than is less than the preset probability; the preset service level satisfied by the expression of the minimum number of battery-swapping robots required by the site is: the average waiting time of users does not exceed the preset time.

In the above technical solution, preferably, the specific process of establishing the user selection behavior model based on the multinomial Logit model includes: assuming that the location of the replacement station has been determined, based on the multinomial Logit model, through the replacement distance and the degree of congestion. Express the utility of the site to be selected; according to the utility maximization principle, express the set of power exchange stations within the maximum power exchange distance that the user is willing to travel; according to the set of power exchange stations and the expression of the utility, the user selects a site to obtain power exchange probability of service.

In the above technical solution, preferably, the specific process of using a distributed robust optimization method to process uncertain parameters and expressing the minimum number of batteries required by a site to meet a preset service level includes: swapping out batteries based on first-in, first-out Rule, on the basis of assuming the number of batteries in the site, express the probability that the SOC of the swapped out battery is not lower than the preset value, and make the probability not less than the preset probability; adopt distributed robust optimization method to deal with the inconsistency in the expression. The parameters are determined and converted to obtain the service level conditional expression to be satisfied by the minimum number of batteries under the user selection decision of the user selection behavior model.

In the above technical solution, preferably, in the process of expressing the minimum number of power exchange robots required by the site under the preset service level: assuming that the power exchange service process conforms to the GI/G/m queuing model, according to the power exchange of the site The average value and variance of the number of robots and the service rate of the battery-swapping robots can be used to obtain the average waiting time for users to obtain battery-swapping services; so that the average waiting time does not exceed the preset time, the conditions for user selection decision-making under a given user selection behavior model Next, get the expression for the minimum number of battery-swapping robots required.

The present invention also proposes a system for robust site selection and capacity determination of a power exchange station considering user selection behavior, applying the robust site selection and capacity determination method for power exchange substations considering user selection behavior as described in any one of the above technical solutions, including: The user selection behavior modeling module is used to establish a user selection behavior model based on multiple Logit models for the behavior of the user selecting a site to obtain power exchange services; the location modeling module is used to construct a location selection based on the user selection behavior model. model; a first constraint transformation module for equivalently transforming the site selection model into a first SOCP constraint; a battery number expression module for processing uncertain parameters by using a distributed robust optimization method, and expressing that a preset service level is satisfied The minimum number of batteries required by the station; the second constraint conversion module is used to equivalently convert the expression of the required minimum number of batteries to form a second SOCP constraint; the battery-swap robot number expression module is used to select the behavior according to the user The user selection decision under the model expresses the minimum number of battery-changing robots required by the site to meet the preset service level; the third constraint transformation module is used to equivalently transform the expression of the minimum number of battery-changing robots required to form the third SOCP Constraints; a modeling module for site selection and capacity, used to combine the first SOCP constraint, the second SOCP constraint, the third SOCP constraint and the objective function of maximizing revenue, to build the distribution station distribution based on the MISOCP model Robust location and capacity model; the solving module is used to call the solver to solve the distributed robust location and capacity model of the battery swap station, and obtain the location and capacity result of the electric vehicle battery swap station.

In the above technical solution, preferably, the preset service level to be satisfied by the expression constructed by the battery number expression module is: the probability that the SOC of the replaced battery is not lower than the preset value is not less than the preset probability; The preset service level to be satisfied by the expression constructed by the electric robot number expression module is: the average waiting time of users does not exceed the preset time.

In the above technical solution, preferably, the specific process for the user selection behavior modeling module to establish a user selection behavior model based on multiple Logit models includes: assuming that the location of the replacement station has been determined, based on the multiple Logit models, by The power exchange distance and the degree of congestion express the utility of the site to be selected; according to the utility maximization principle, express the set of power exchange stations within the maximum power exchange distance that the user is willing to travel; obtain the user selection according to the set of power exchange stations and the expression of the utility The probability that a site will obtain a battery swap service.

In the above technical solution, preferably, the battery number expression module adopts a distributed robust optimization method to process uncertain parameters, and the specific process of expressing the minimum number of batteries required by the site to meet the preset service level includes: based on a first-in, first-out basis Based on the battery swapping rule, the probability that the SOC of the swapped battery is not lower than the preset value is expressed on the basis of assuming the number of batteries in the site, and the probability is not less than the preset probability; the distributed robust optimization method is used to process the expression The uncertain parameters in the formula are converted to obtain the expression of the minimum number of batteries required to meet a certain service level under the user selection decision of the user selection behavior model.

In the above technical solution, preferably, the process of expressing the minimum number of battery-changing robots required by the site to meet the preset service level by the battery-changing robot number expression module specifically includes: assuming that the battery-changing service process conforms to GI/G/m queuing Model, according to the number of battery-changing robots at the site and the mean and variance of the service rate of the battery-changing robots, the average waiting time for users to obtain battery-changing services is obtained; so that the average waiting time does not exceed the preset time, in a given user selection behavior model Under the condition of the user's choice decision under , the expression of the required minimum number of battery-swapping robots is obtained.

Compared with the prior art, the beneficial effects of the present invention are: by establishing a user selection behavior model, a site selection model, and constructing the minimum number of batteries and the minimum number of battery-changing robots required for the site under the condition of satisfying the preset service level, Under the objective function of maximizing revenue, a robust distributed location and capacity model for battery swapping stations is constructed and equivalently converted into a MISOCP model, so as to obtain the results of location and capacity determination of electric vehicle battery swapping stations. The method provided by the invention comprehensively considers uncertainty and user selection behavior, and at the same time, it can maximize the profit of the operator's site selection and construction, battery and battery replacement robots, and improve the scientificity of site selection and construction and capacity determination.

Description of drawings

FIG. 1 is a schematic flowchart of a method for robust site selection and capacity determination of a power exchange station considering user selection behavior disclosed by an embodiment of the present invention;

FIG. 2 is a schematic structural block diagram of a system for robust site selection and capacity determination of a power exchange station that considers user selection behavior disclosed in an embodiment of the present invention.

In the figure, the corresponding relationship between each component and the reference sign is:

11. User selection behavior modeling module, 12. Location modeling module, 13. First constraint transformation module, 14. Battery number expression module, 15. Second constraint transformation module, 16. Battery replacement robot number expression module, 17 . The third constraint transformation module, 18. The modeling module of location selection and volume, 19. The solving module.

Detailed ways

In order to make the purposes, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments These are some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present invention.

Below in conjunction with accompanying drawing, the present invention is described in further detail:

As shown in FIG. 1 , according to a method for robust site selection and capacity determination of a power exchange station considering user selection behavior provided by the present invention, the method includes: establishing user selection behavior based on multiple Logit models for the behavior of user selection of a site to obtain power exchange services Model; build a site selection model based on the user selection behavior model; convert the site selection model equivalently to form the first SOCP constraint; use a distributed robust optimization method to deal with uncertain parameters, and express the minimum required by the site to meet the preset service level Number of batteries; equivalently transform the expression of the minimum number of batteries required to form the second SOCP constraint; according to the user selection decision under the user selection behavior model, express the minimum number of battery-changing robots required by the site to meet the preset service level; The expression of the minimum number of battery-swapping robots required is equivalently transformed to form the third SOCP constraint; combined with the first SOCP constraint, the second SOCP constraint, the third SOCP constraint and the objective function of maximizing revenue, the swapping station is constructed based on the MISOCP model Distributed robust location and capacity model; call the solver to solve the distributed robust location and capacity model of the battery swap station, and obtain the location and capacity result of the electric vehicle battery swap station.

In this embodiment, unlike the route-based power exchange demand on the inter-city high-speed transportation network, the intra-city power exchange demand is a node-based demand. For example, electric taxi drivers in cities often need to refuel after multiple trips. After the customer gets off the bus, the driver can select one of the stations that can provide the battery replacement service based on the current location information and personal preference to obtain the battery replacement service, that is, there is a user selection behavior. In this embodiment, considering the user's selection behavior, the uncertainty of the power exchange demand and the uncertainty of the SOC of the replaced battery, firstly, a multinomial Logit model is used to model the user's selection behavior, and the distributed robust optimization method is used to deal with the uncertainty parameters, and construct a distributed robust site selection and capacity model for battery swap stations that satisfies a certain service level. Second, the model is equivalently transformed into a MISOCP model.

Unlike the energy replenishment method of private cars, taxi drivers usually replenish energy after they have delivered passengers to a designated location. For each driver, the choice of swap stations when the customer gets off the bus is usually not unique. The driver will select a certain power exchange station with a certain probability to exchange electricity according to certain factors and personal preference. From the perspective of the operator, the present invention considers that in the existing candidate site selection nodes of the power exchange station in the city, some site selection nodes are selected to build the station, and a certain number of them are placed under the condition that the power exchange demand and the SOC of the replaced battery are uncertain. of batteries and power-swapping robots, maximizing the operator's profit.

Different from the charging station, a certain number of batteries and battery swapping robots need to be configured in the swapping station. Too many batteries will increase inventory holding costs, while too few batteries will require users to wait longer for higher-charged batteries. Too many power-changing robots will increase investment costs, while too few robots will increase the waiting time of users in line for power-changing. Therefore, the service level to be satisfied by the site selection and capacity determination of the swapping station is defined from two aspects: for the minimum number of batteries required by the site, the SOC of the swapped batteries is guaranteed not to be lower than a certain value with a certain probability. The minimum number of power exchange robots required by the site, and the average waiting time of users does not exceed a certain value.

Let Q be the set of drop-off points for electric taxi customers, and N be the set of swap stations that can serve electric taxis. Define b ₁ , b ₂ , and b ₃ as the unit price of battery purchase, the unit price of power exchange robot, and the unit price of power exchange income. The goal of this model is to determine the location and capacity of the swap station, so as to maximize the time-averaged profit of the swap station.

Three types of decision variables are defined:

y _i : build a station at point i, the value is 1, i∈N; otherwise, the value is 0

w _i : the number of battery swapping robots purchased at the point, i∈N

m _i : power exchange demand at point, i∈N

In view of the above problems, the following model is established with the maximization of income as the objective function:

π∈Π(y),w∈W(y,π),m∈M(y,π) (5)

Among them, formula (1) is the objective function, the first item is the power exchange income, the second item is the investment and operation cost of the station construction, the third item is the battery purchase cost, and the fourth item is the power exchange robot purchase cost; constraints (2) and ( 3) is the total cost of battery purchase and the total cost of power exchange robot; constraints (4)-(5) define the feasible domain of decision variables, where Π(y) is the power exchange feasible domain of the power exchange station, which is determined by the location selection y is determined, and W(y, π) and M(y, π) are the feasible regions for the number of batteries purchased by the swap station, which are both determined by the location decision y and the power swap demand π.

In the above embodiment, preferably, assuming that the location of the swapping station has been determined, each user can choose whether to swap electricity and which swapping station to go to according to their own needs and personal preferences. The main factors that affect the user's selection behavior of the swapping station are: the distance to the swapping station, other service facilities near the swapping station, and the number of service facilities at the swapping station, and so on.

Based on the multinomial Logit model, the utility of the station to be selected is expressed by the distance of power exchange and the degree of congestion; according to the principle of maximization of utility, the set of power exchange stations within the maximum power exchange distance that the user is willing to travel is expressed; according to the expression of the set of power exchange stations and the utility Obtain the probability that a user selects a site to obtain a battery swap service.

Specifically, the process of building a user selection behavior model is as follows:

The utility U _qi of power exchange from the demand at point q to point i is composed of the deterministic utility V _qi and the uncertainty utility ε _qi , namely

U _qi =V _qi +ε _qi , (6)

用q点到i点的距离d_qi表示换电距离，用预测的i点的平均换电需求表示该点的拥挤程度

假设V_qi是关于d_qi与

的线性函数，并且V_qi关于d_qi和

均为负相关。假设Among them, the determination of the utility V _qi is determined by two parts: the replacement distance d _qi and the degree of crowding at point i

The distance d _qi from point q to point i is used to represent the distance of power exchange, and the predicted average power exchange demand of point i is used to represent the congestion degree of the point.

Suppose V _qi is about d _qi and

A linear function of , and V _qi with respect to d _qi and

are negatively correlated. Assumption

的权重系数。MNL中的随机效用ε_qi是独立同分布的。where β ₀ and β ₁ are d _qi and

weight factor. The random utility _εqi in MNL is independent and identically distributed.

After the site selection decision is given, the user will select the swap station according to the principle of utility maximization. N _q is used to represent the set of substations that can serve the demand of point q (q∈Q), which consists of points whose distance from point q does not exceed d _q,max , that is, N _q ={i∈N:d _qi ≤d _q,max }, where d _q,max is the maximum battery swap distance that the user at point q is willing to travel. In the actual situation, the driver may choose not to change the battery due to reasons such as distance or time. We use i=0 to represent the situation where the user does not change the battery. According to the user selection behavior model, the probability P _qi of the user at point q to exchange power at the power exchange station i is:

Simplify it to get

Without loss of generality, still use V _qi to represent V _qi -V _q0 , we can get

则q点处的需求到i点换电的概率p_qi为Based on the above user selection behavior model, the next step is to model the site selection problem of the battery swapping station. This problem can be regarded as a two-stage problem. In the first stage, the operator decides the location of the swapping station. In the second stage, the location of the swapping station is determined, and the user selects the swapping station to obtain the battery swapping service. make

Then the probability p _qi of the demand at point q to change power at point i is:

When i=0, that is, when the user does not change the battery, there are a _q0 =1, y ₀ =1. Therefore, the probability p _qi of the demand at point q to exchange electricity at point i is

Next, the above site selection model is transformed into SOCP constraints:

因此可以将

松弛为

make

Therefore, it can be

relax to

is equivalent to the following constraints:

Therefore the feasible region of p _qi is defined by the following SOCP constraints:

In the above embodiment, preferably, based on the first-in-first-out battery swapping rule, the probability that the SOC of the swapped battery is not lower than the preset value is expressed on the basis of assuming the number of batteries in the site, and the probability is not less than the preset value. Probability; using distributed robust optimization method to deal with uncertain parameters in the expression, transform to obtain the expression of the minimum number of batteries required to meet a certain service level under the user selection decision of the user selection behavior model, and convert it into SOCP constraint.

Specifically, it is assumed that there are n batteries in a power exchange station, and the power exchange requests arrive randomly, let T be the time interval between two randomly arrived power exchange requests, X is the SOC of the swapped-in battery, and Z is the swapped-out battery. SOC, X and Z are all random variables.

Given the target value of the swapped-out battery SOC θ∈(0,1), the service level related to the swapped-out battery SOC is defined as Prob (Z≥θ). Assume that the battery swap rule is First In First Out (FIFO), that is, the battery that arrives first is charged first and swapped out first, so that the probability that the SOC of the swapped battery is not lower than θ is not lower than a certain value, That is, it must be guaranteed that Prob(Z≥θ)≥1-∈, where ∈∈(0,1).

In real life, the exact probability distribution functions of T and X are unknown, but their mean and variance information can be known from historical data. Therefore, the present invention adopts a distributed robust optimization method based on moment information to deal with uncertain parameters.

因此，对于任何一个凸集S，有According to the following lemma: Suppose

Therefore, for any convex set S, we have

假设随机变量Z的期望为

方差为Var[Z]＝σ²＞0。令S＝{z:z＜θ}。因为in,

Suppose the expectation of the random variable Z is

The variance is Var[Z]=σ ² >0. Let S={z:z<θ}. because

so

Three assumptions are made here:

1. Charge unit time, the battery SOC increases by 1% (if not fully charged).

2.

3. _Ti and Xi are _irrelevant .

Let Y=(T ₁ ,...,T _n ,X ₁ ) ^T , μ=(μ _T ,...,μ _T ,μ _X ) ^T , and Σ is the covariance matrix of Y. Based on the above assumptions, under the FIFO battery swap strategy, the SOC of the swapped out battery is

Consider the following questions:

Equivalent to

because

to ensure that

must make θ≤e ^T μθ≤e ^T μ,

because

故，FIFO换电策略下所需的最少电池数应该满足：

Therefore, the minimum number of batteries required under the FIFO power swap strategy should satisfy:

Suppose θ ≥ μ _X . make

可以得到according to

can get

因此，(23)式中第一个约束是多余的。因此，FIFO策略下所需的最少电池数由(24)式给出。As can be seen,

Therefore, the first constraint in (23) is redundant. Therefore, the minimum number of batteries required under the FIFO strategy is given by (24).

为f_q的均值与方差，并且假设{f_q,q∈Q}是相互独立的。因此，换电站i处的换电需求为

Let f _q be the random power exchange demand at the demand point q∈Q, let

are the mean and variance of f _q , and assume that {f _q ,q∈Q} are independent of each other. Therefore, the power exchange demand at the power exchange station i is

make

According to the nature of the update process, for a given decision p, the mean and covariance of random time intervals between two consecutively arriving EVs at node i are given as

Under the FIFO strategy, according to equation (25), the minimum number of batteries required at node i is

Next, transform this expression to form a SOCP constraint:

是关于

的单调递增函数。用决策变量w_i表示i点处所需的电池数。Obviously,

its about

a monotonically increasing function. The number of batteries required at point _i is represented by the decision variable wi.

make

but

松弛为

Since w _i ≥h(π _i ,τ _i ,vi ₎ increases monotonically with respect to τ _i and v _i . can be

relax to

is equivalent to the following constraints:

Finally, by introducing the variable _ui , we can get

So the feasible region of _wi is defined by the following SOCP constraints:

In the above embodiment, preferably, it is assumed that the battery swapping service process conforms to the GI/G/m queuing model, and the average waiting time of the user to obtain the battery swapping service is obtained according to the number of battery swapping robots at the site and the mean and variance of the battery swapping robot service rate. time; so that the average waiting time does not exceed the preset time, under the condition of the user's choice decision under the user's choice behavior model, the expression of the required minimum number of battery-swapping robots is obtained, and it is converted into SOCP constraints.

假设到达率的均值和方差为μ_r，

令c_s＝σ_s/μ_s，c_T＝σ_T/μ_T，其中，μ_T＝1/μ_r，

Specifically, it is assumed that there are m battery-changing robots in the battery-changing station, and the mean and variance of the service rate of a single battery-changing robot is μ _s ,

Assuming that the mean and variance of the arrival rate are μ _r ,

Let c _s =σ _s /μ _s , c _T =σ _T /μ _T , where μ _T =1/μ _r ,

The power exchange process of each station can be simulated as a GI/G/m queuing model, that is, general arrival, general service, and multiple service desks. According to the Allen-Cunneen approximation principle, the average waiting time of users is

where ρ=μ _r /(mμ _s ).

In order to ensure that the average waiting time of users does not exceed L, it is necessary to make

which is

因此，可以得到in,

Therefore, it can be obtained

Next, transform this expression to form a SOCP constraint:

For each point i, given the user selection decision p, we have

The above constraints are equivalent to

in,

Using the notation in (31), the above constraints can be written as

According to the three SOCP constraints formed above, the objective function of maximizing revenue and the feasible region of each decision variable, a robust distributed location selection and capacity model for battery swap stations based on the MISOCP model is constructed as follows:

The solver is called to directly solve the distributed robust location and capacity model of the battery swap station, and the results of the location and capacity determination of the electric vehicle battery swap station are obtained.

The present invention also proposes a system for robust site selection and capacity determination of a power exchange station considering user selection behavior, applying the robust site selection and capacity determination method for power exchange substations considering user selection behavior proposed in any of the foregoing embodiments, including:

The user selection behavior modeling module 11 is used to establish a user selection behavior model based on multiple Logit models for the behavior of the user selecting the site to obtain the battery swap service;

The site selection modeling module 12 is used for constructing a site selection model based on the user selection behavior model;

The first constraint conversion module 13 is used to equivalently convert the site selection model to form the first SOCP constraint;

The battery number expression module 14 is used for using the distributed robust optimization method to process the uncertain parameters, and expressing the minimum number of batteries required by the site to meet the preset service level;

The second constraint conversion module 15 is configured to equivalently convert the expression of the required minimum number of batteries to form a second SOCP constraint;

The number of battery-changing robots expression module 16 is used to express the minimum number of battery-changing robots required by the site under the preset service level according to the user's selection decision under the user's selection behavior model;

The third constraint conversion module 17 is used to equivalently convert the expression of the required minimum number of battery-swapping robots to form a third SOCP constraint;

The site selection and capacity modeling module 18 is used to combine the first SOCP constraint, the second SOCP constraint, the third SOCP constraint and the objective function of maximizing revenue to construct a distributed robust site selection and capacity model for power swap stations based on the MISOCP model ;

The solving module 19 is used for invoking the solver to solve the distributed robust location and capacity model of the battery swap station to obtain the location and capacity result of the electric vehicle battery swap station.

In the above embodiment, preferably, the preset service level to be satisfied by the expression constructed by the battery number expression module 14 is: the probability that the SOC of the swapped battery is not lower than the preset value is not less than the preset probability; The preset service level to be satisfied by the expression constructed by the expression module 16 is: the average waiting time of users does not exceed the preset time.

In the above embodiment, preferably, the specific process that the user selection behavior modeling module 11 establishes the user selection behavior model based on the multinomial Logit models includes: assuming that the location of the replacement power station has been determined, based on the multinomial Logit models, through the replacement Electric distance and congestion degree express the utility of the site to be selected; according to the principle of maximization of utility, express the set of power exchange stations within the maximum power exchange distance that the user is willing to travel; according to the expression of the set of power exchange stations and the utility, the user selects a site to obtain the exchange station. Probability of Electric Service.

In the above embodiment, preferably, the battery number expression module 14 adopts a distributed robust optimization method to process uncertain parameters, and the specific process of expressing the minimum number of batteries required by a site to meet the preset service level includes: based on a first-in, first-out The battery swapping rule expresses the probability that the SOC of the swapped battery is not lower than the preset value on the basis of assuming the number of batteries in the site, and makes the probability not less than the preset probability; the distributed robust optimization method is used to process the SOC in the expression. Uncertain parameters, the conversion obtains the expression of the minimum number of batteries required to meet a certain service level under the user selection decision of the user selection behavior model.

In the above embodiment, preferably, the process of expressing the minimum number of battery-changing robots required by the site under the preset service level by the battery-changing robot number expression module 16 specifically includes: assuming that the battery-changing service process conforms to the GI/G/m queuing model , according to the number of battery-changing robots at the site and the mean and variance of the service rate of battery-changing robots, the average waiting time for users to obtain battery-changing services is obtained; so that the average waiting time does not exceed the preset time, users under a given user selection behavior model Under the condition of selection decision, get the expression of the minimum number of battery-swapping robots required.

The above are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims (10)

1. A robust site selection and volume fixing method for a power station considering user selection behaviors is characterized by comprising the following steps:

aiming at the behavior of acquiring the battery swapping service by the user selection site, establishing a user selection behavior model based on a plurality of Logit models;

constructing an addressing model based on the user selection behavior model;

equivalently converting the addressing model to form a first SOCP constraint;

processing uncertain parameters by adopting a distributed robust optimization method, and expressing the minimum number of batteries required by the station under the condition of meeting the preset service level;

equivalently converting the expression of the required minimum battery number to form a second SOCP constraint;

expressing the minimum number of the electricity changing machines required by the station under the preset service level according to the user selection decision under the user selection behavior model;

equivalently converting the expression of the minimum number of the battery replacement robots to form a third SOCP constraint;

combining the first SOCP constraint, the second SOCP constraint, the third SOCP constraint and an objective function of profit maximization, and constructing a power station distribution robust siting constant volume model based on a mixed integer second-order cone programming model;

and calling a solver to solve the power station distribution robust site selection constant volume model to obtain a site selection constant volume result of the electric automobile power station.

2. The robust site selection sizing method considering the user selection behavior for the power change station as claimed in claim 1, wherein the expression of the minimum number of batteries required by the station satisfies a preset service level of: the probability that the state of charge of the battery to be exchanged is not lower than the preset value is not lower than the preset probability;

the expression of the minimum number of the power change machines required by the station satisfies the preset service level as follows: the average waiting time of the users does not exceed the preset time.

3. The robust siting capacity method for a power conversion station considering user selection behavior as claimed in claim 1, wherein the specific process of establishing the user selection behavior model based on the multiple Logit models comprises:

under the condition that the site selection of the power swapping station is determined, the utility of the station to be selected is expressed through the power swapping distance and the crowding degree based on a plurality of Logit models;

expressing a power change station set within the maximum power change distance which a user is willing to travel according to a utility maximization principle;

and obtaining the probability that a user selects a certain station to obtain the battery swapping service according to the battery swapping station set and the utility expression.

4. The robust power station location determination method considering the user selection behavior as claimed in claim 2, wherein the specific process of processing the uncertain parameters by using the distributed robust optimization method and expressing the minimum number of batteries required by the station satisfying the preset service level comprises:

based on a first-in first-out battery swap-out rule, expressing the probability that the state of charge of a swapped-out battery is not lower than a preset value on the basis of the assumed battery number in a station, and enabling the probability not to be lower than the preset probability;

and processing uncertain parameters in the expression by adopting a distributed robust optimization method, and converting to obtain the expression of the minimum battery number required by meeting a certain service level under the user selection decision of the user selection behavior model.

5. The robust siting and sizing method for a power swapping station considering user selection behavior as claimed in claim 2, wherein in the process of expressing the minimum number of power swapping machines required by the station under a preset service level:

assuming that the battery swapping service process conforms to a GI/G/m queuing model, and obtaining the average waiting time of the user for acquiring the battery swapping service according to the number of battery swapping robots at the station and the mean and variance of the service rate of the battery swapping robots;

and obtaining the expression of the minimum number of the battery replacement robots under the condition of the user selection decision given under the user selection behavior model, wherein the average waiting time is not more than the preset time.

6. A robust siting system for a power switching station considering user selection behaviors, which applies the robust siting method for the power switching station considering user selection behaviors as claimed in any one of claims 1 to 5, and is characterized by comprising the following steps:

the user selection behavior modeling module is used for establishing a user selection behavior model based on a plurality of Logit models aiming at the behavior of obtaining the battery swapping service by the user selecting the site;

the address selection modeling module is used for constructing an address selection model based on the user selection behavior model;

the first constraint conversion module is used for equivalently converting the addressing model to form a first SOCP constraint;

the battery number expression module is used for processing the uncertain parameters by adopting a distributed robust optimization method and expressing the minimum battery number required by the station under the condition of meeting the preset service level;

the second constraint conversion module is used for equivalently converting the expression of the required minimum battery number to form a second SOCP constraint;

the number of the power conversion robots is used for expressing the minimum number of the power conversion robots required by the station under the preset service level according to the user selection decision under the user selection behavior model;

the third constraint conversion module is used for equivalently converting the expression of the minimum number of the battery swapping robots to form a third SOCP constraint;

the locating and sizing modeling module is used for constructing a power station distribution robust locating and sizing model on the basis of a mixed integer second-order cone programming model by combining the first SOCP constraint, the second SOCP constraint, the third SOCP constraint and an objective function of profit maximization;

and the solving module is used for calling a solver to solve the power station distribution robust site selection constant volume model to obtain a site selection constant volume result of the electric automobile power station.

7. The robust siting capacity system for a power conversion station considering user selection behavior as claimed in claim 6, wherein the preset service level to be satisfied by the expression constructed by the battery count expression module is: the probability that the state of charge of the battery to be exchanged is not lower than the preset value is not lower than the preset probability;

the preset service level to be met by the expression constructed by the number of the battery swapping machines is as follows: the average waiting time of the users does not exceed the preset time.

8. The robust siting capacity system for a power conversion station considering user selection behavior as claimed in claim 6, wherein the specific process of the user selection behavior modeling module establishing the user selection behavior model based on the plurality of Logit models comprises:

under the condition that the site selection of the power swapping station is determined, the utility of the station to be selected is expressed through the power swapping distance and the crowding degree based on a plurality of Logit models;

expressing a power change station set within the maximum power change distance which a user is willing to travel according to a utility maximization principle;

and obtaining the probability that a user selects a certain station to obtain the battery swapping service according to the battery swapping station set and the utility expression.

9. The robust power station location determination system considering the user selection behavior as claimed in claim 7, wherein the specific process of the battery number expression module processing the uncertain parameters by using a distributed robust optimization method and expressing the minimum number of batteries required by the station satisfying the preset service level comprises:

based on a first-in first-out battery swap-out rule, expressing the probability that the state of charge of a swapped-out battery is not lower than a preset value on the basis of the assumed battery number in a station, and enabling the probability not to be lower than the preset probability;

and processing uncertain parameters in the expression by adopting a distributed robust optimization method, and converting to obtain a service level conditional expression which is required to be met by the minimum battery number under the user selection decision of the user selection behavior model.

10. The robust siting capacity system for a power swapping station considering user selection behavior as claimed in claim 7, wherein the process of the number of power swapping robot expression module expressing the minimum number of power swapping robot required by the station meeting the preset service level specifically comprises:

assuming that the battery swapping service process conforms to a GI/G/m queuing model, and obtaining the average waiting time of the user for acquiring the battery swapping service according to the number of battery swapping robots at the station and the mean and variance of the service rate of the battery swapping robots;

and obtaining the expression of the minimum number of the battery replacement robots under the condition of the user selection decision given under the user selection behavior model, wherein the average waiting time is not more than the preset time.

Images