CN109800917A

CN109800917A - A kind of planing method in electric car parking lot, device and calculate equipment

Info

Publication number: CN109800917A
Application number: CN201910065012.5A
Authority: CN
Inventors: 曾博; 刘裕; 徐富强; 卫璇; 冯家欢; 朱志伟
Original assignee: North China Electric Power University
Current assignee: North China Electric Power University
Priority date: 2019-01-23
Filing date: 2019-01-23
Publication date: 2019-05-24

Abstract

The invention discloses a kind of planing methods in electric car parking lot, suitable for being executed in calculating equipment, this method comprises: building degree of regretting Mechanism Model, and the automobile user model based on decision-making uncertainty is constructed according to degree of the regretting Mechanism Model, which includes that decision relies on utility function；Pass through the uncertain uncertain model of place for carrying out clustering and constructing electric car to electric car；According to the two-stage programming model of automobile user model and uncertain model of place building electric car parking lot, which includes lower layer's planning of parking lot addressing, constant volume, the upper layer planning of price incentive design and parking lot traffic control；And pre-defined algorithm is respectively adopted, the upper layer and lower layer planning of two-stage programming model is solved, obtain the optimum programming scheme in electric car parking lot.Calculating equipment the invention also discloses the device for planning in corresponding electric car parking lot and for executing this method.

Description

Planning method and device for electric vehicle parking lot and computing equipment

Technical Field

The invention relates to the field of power systems, in particular to a planning method and device for an electric automobile parking lot and computing equipment.

Background

In order to meet the access requirements of large-scale electric automobiles on an electric power system, a large number of electric automobile charging facilities are also planned along residential areas, shopping malls and expressways, and different types of electric automobiles can be charged on different charging facilities. When electric automobile and big electric wire netting are carrying out charge-discharge on filling electric pile, the electric power demand of charging has influenced the load space-time distribution of big electric wire netting, also provides the energy storage of a large amount of removals for electric power system. Just as private electric vehicles can be charged on a slow charging pile at home and also on a fast charging pile of the electric vehicle, different types of charging facilities can meet the charging requirements of different types of users, wherein a public electric vehicle parking lot is one of the most concerned charging media as the best media for electric vehicle integrators to interact with the users. Different from a quick charging station mainly used for emergency charging, the bus parking lot has the advantage that the electric automobile stays in the bus parking lot for a longer time due to the parking function, so that the possibility is provided for the automobile parked in the bus parking lot to participate in demand side response, and the bus parking lot becomes a medium for interaction of the electric automobile and a large power grid. Planning and deploying appropriate public charging facilities, namely, controlling the electric vehicle access to the power system through an intelligent electric vehicle parking lot is a good method.

Meanwhile, in the real world, most electric vehicles are private, each electric vehicle user can charge the electric vehicle in different modes, and whether the user wants to enter and use the electric vehicle parking lot for a long time becomes a key factor for determining investment efficiency. The rapid development of the behavior economics enables people to have a more pertinent description mode for behavior selection of electric automobile users. The charging behavior of an electric vehicle is not only dependent on the personal preferences of the vehicle owner, but is also influenced by the geographical conditions of the charging facility and by monetary subsidies. In practical research, as the investment decision and the subsidy policy of the intelligent parking lot are decision variables and influence on the willingness of the charging behavior of the electric automobile, the participation willingness of the electric automobile user depends on the decision variables and cannot be represented by a given probability distribution, so that the decision-dependent uncertainty modeling is used for representing the probability distribution of the participation willingness of the user as a function of the previous subsidy and the investment decision, and the user behavior is more accurately depicted. The utilization rate of the intelligent parking lot charging pile is likely to be estimated wrongly according to the planning and subsidy policy making result without considering the user intention, so that non-optimal decision is made. Therefore, the intelligent parking lot planning research considering decision dependence uncertainty has practical significance.

Disclosure of Invention

To this end, the present invention provides a planning, arrangement and computing device for an electric car park in an attempt to solve or at least alleviate the problems presented above.

According to an aspect of the present invention, there is provided a method for planning a parking lot for an electric vehicle, adapted to be executed in a computing device, the method comprising: constructing a repentance mechanism model according to a reflection-reaction paradigm, and constructing an electric vehicle user model based on decision-dependent uncertainty according to the repentance mechanism model, wherein the electric vehicle user model comprises a decision-dependent utility function; constructing an uncertainty scene model of the electric vehicle by performing cluster analysis on uncertainty of the electric vehicle, wherein the uncertainty scene model comprises all states of the electric vehicle parking lot which can occur during operation; constructing a two-stage planning model of the electric automobile parking lot according to the electric automobile user model and the uncertainty scene model, wherein the two-stage planning model aims at obtaining the maximum profit of an electric automobile integrator in the power distribution network and comprises upper-layer planning of parking lot site selection, volume fixing and price incentive design and lower-layer planning of parking lot operation scheduling; and respectively solving the upper-layer plan and the lower-layer plan of the two-stage planning model by adopting a predetermined algorithm to obtain an optimal planning scheme of the electric automobile parking lot.

Optionally, in the planning method according to the present invention, the decision-dependent utility function includes:

wherein, W_y,iIs the benefit of the electric vehicle i at the time interval y,is the reward and subsidy for the electric vehicle i at time interval y,is the inconvenient cost of the electric automobile i at the time interval y,is the battery charge-discharge loss cost of the electric vehicle i at the time interval y.

Optionally, in the planning method according to the present invention, the uncertainty of the electric vehicle includes an endogenous uncertainty and an exogenous uncertainty, wherein the endogenous uncertainty includes an electric vehicle engagement factor, and the exogenous uncertainty includes an initial state of charge, an arrival time, and a departure time of the electric vehicle.

Optionally, in the planning method according to the present invention, the upper-layer planning algorithm in the first stage is a genetic algorithm, and the lower-layer planning algorithm in the second stage is a meta-dual interior point method.

Optionally, in the planning method according to the present invention, the upper objective function of the two-phase planning model is profit F of the electric vehicle integrator in the power distribution network^PLMaximization, which is calculated by the formula:

wherein, Λ^OpeAnnual operating income, C^InvIs the annual equivalent investment cost,is the number of electric vehicle charging stations set at node b,is a binary variable, omega, representing whether the electric car park is established at node b_SIs a set of scenes, Y is a contract period of one year, ρ_y,sIs the probability of the scene s occurring at the time interval y, Λ_y,sIs the operating revenue, k, of scene s at time interval y^cpIs a year-valued operator, k, of the charging pile^ldIs a annual value operator of the land, pi^cpIs the investment cost of the bidirectional charger.

Optionally, in the planning method according to the present invention, the upper constraint conditions of the two-phase planning model are:

wherein,is the maximum number of electric vehicle charging stations, Rew, set at node b_bThe incentive subsidy price is set at node b,is the maximum incentive subsidy price set at node b.

Optionally, in the planning method according to the present invention, the lower objective function of the two-stage planning model maximizes the operating profit of the parking lot, and the decision variables include the charge and discharge power of the electric vehicle parking lot and a binary variable of whether the user signs a contract with the electric vehicle parking lot.

Optionally, in the planning method according to the present invention, the lower objective function of the two-phase planning model is:

wherein,represents the operation income of the electric automobile parking lot,represents the running cost of the electric car parking lot,represents the contract customization cost of the electric automobile parking lot, theta represents the number of days in each operation period, omega_IIs a set of users that are in a group,represents the discharge power of the electric car parking lot,represents the discharge electricity rate of the electric vehicle, r is an operation time interval,a binary variable indicating whether or not the electric vehicle user enters a price incentive contract with the s parking lot,represents the charging cost of the electric vehicle, pi^omRepresents the daily maintenance cost of the electric vehicle,represents the charging power of the electric car parking lot,indicating the real-time electricity price, Rew_bRepresenting the incentive fee signed by the electric automobile parking lot and the electric automobile user.

Optionally, in the planning method according to the present invention, the lower layer constraint condition of the two-stage planning model includes at least one of a maximum charge-discharge power constraint, a constraint that charge and discharge cannot be performed simultaneously, an electric vehicle state of charge constraint, a constraint that electric vehicle charging demand is satisfied, an electric vehicle battery loss constraint, an electric vehicle available binary constraint, an electric vehicle contract amount constraint, an electric vehicle available amount constraint, an electric vehicle arrival amount constraint, and an electric vehicle departure amount constraint.

Optionally, in the planning method according to the present invention, the maximum charge-discharge power constraint is:

wherein, γ^maxRepresents the maximum charge and discharge power of the charging pile of the electric automobile,representing the number of dispatchable electric vehicles in the electric vehicle parking lot, T being the period in the time period T of the day.

Optionally, in the planning method according to the present invention, the constraint that charging and discharging cannot be performed simultaneously is:

optionally, in the planning method according to the present invention, the electric vehicle state of charge constraint is:

wherein E is_y,s,b,tRepresenting the total state of charge of the electric vehicle at the current stage, E_y,s,b,t-1Representing the total state of charge of the electric vehicle in the previous stage, η representing the charge-discharge efficiency,represents the energy capacity of the battery of the electric automobile,represents the state of charge of the electric vehicle when arriving,represents the state of charge that the electric automobile needs to reach,indicating the number of arriving electric vehicles in the electric vehicle parking lot,indicating the number of exiting electric vehicles in the electric vehicle parking lot.

Optionally, in the planning method according to the present invention, the electric vehicle battery loss constraint is:

wherein, pi^dgIndicating the battery loss cost of the electric automobile, and psi indicating the battery loss of the electric automobile parking lotThe consumption is limited to a constant value that is,representing the nominal charging power in the normal mode.

According to another aspect of the present invention, there is provided an electric vehicle parking lot planning apparatus adapted to be resident in a computing device for execution, the apparatus comprising: the user model building unit is suitable for building a regret degree mechanism model according to a reflection-reaction paradigm, and building an electric vehicle user model based on decision-dependent uncertainty according to the regret degree mechanism model, wherein the electric vehicle user model comprises a decision-dependent utility function; the scene model building unit is suitable for building an uncertainty scene model of the electric automobile through clustering analysis on uncertainty of the electric automobile, wherein the uncertainty scene model comprises all states which can occur in an electric automobile parking lot during operation; the planning model construction unit is suitable for constructing a two-stage planning model of the electric automobile parking lot according to an electric automobile user model and an uncertainty scene model, the two-stage planning model takes the maximum profit obtained by an electric automobile integrator in a power distribution network as a target, and the two-stage planning model comprises upper-layer planning of parking lot site selection, constant volume and price incentive design and lower-layer planning of parking lot operation scheduling; and the model solving unit is suitable for respectively adopting a predetermined algorithm to solve the upper-layer plan and the lower-layer plan of the two-stage planning model so as to obtain the optimal planning scheme of the electric automobile parking lot.

According to yet another aspect of the present invention, there is provided a computing device comprising: one or more processors; a memory; and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for performing the method of planning an electric vehicle parking lot as described above.

According to yet another aspect of the present invention, there is provided a readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by a computing device, cause the computing device to perform the method of planning an electric vehicle parking lot as described above.

According to the technical scheme, the probability density function is used for describing the exogenous uncertainty of the electric automobile aiming at a single electric automobile, and meanwhile, the influence of the orderly charging of the electric automobile on a power distribution system is simulated based on the concept of a virtual power plant. Aiming at the endogenous uncertainty of the electric automobile, namely the investment decision of the current stage possibly has great influence on the space-time distribution of the charging load of the electric automobile in the future, a model for describing the decision-dependent uncertainty of the electric automobile is constructed. The invention constructs a regret degree mechanism model according to a reflection-reaction paradigm to calculate the probability distribution of behaviors of electric vehicle users under different decisions and contract incentives. In addition, a utility function of the quantitative regret degree is described in detail, and finally, an uncertainty scene is generated by utilizing cluster analysis, so that modeling of the whole decision-dependent uncertainty is completed. On the basis of the uncertainty modeling and the statistical scene modeling, the method takes the maximum profit obtained by an electric vehicle integrator in the power distribution network as an objective function, makes all planning decisions and encourages contract design, and establishes an electric vehicle parking lot planning model based on decision-dependent uncertainty.

Drawings

To the accomplishment of the foregoing and related ends, certain illustrative aspects are described herein in connection with the following description and the annexed drawings, which are indicative of various ways in which the principles disclosed herein may be practiced, and all aspects and equivalents thereof are intended to be within the scope of the claimed subject matter. The above and other objects, features and advantages of the present disclosure will become more apparent from the following detailed description read in conjunction with the accompanying drawings. Throughout this disclosure, like reference numerals generally refer to like parts or elements.

FIG. 1 shows a schematic diagram of a computing device 100, according to one embodiment of the invention;

FIG. 2 shows a flow diagram of a method 200 of planning a parking lot for an electric vehicle according to one embodiment of the invention;

fig. 3 is a block diagram illustrating a configuration of an electric car parking lot planning apparatus 300 according to an embodiment of the present invention;

FIG. 4 illustrates a framework diagram of a two-phase planning model, according to one embodiment of the invention;

FIG. 5 illustrates a two-phase planning model solution diagram according to one embodiment of the invention; and

fig. 6 shows a schematic diagram of an IEEE12 node system according to one embodiment of the invention.

Detailed Description

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

Fig. 1 is a block diagram of an example computing device 100. In a basic configuration 102, computing device 100 typically includes system memory 106 and one or more processors 104. A memory bus 108 may be used for communication between the processor 104 and the system memory 106.

Depending on the desired configuration, the processor 104 may be any type of processing, including but not limited to: a microprocessor (μ P), a microcontroller (μ C), a Digital Signal Processor (DSP), or any combination thereof. The processor 104 may include one or more levels of cache, such as a level one cache 110 and a level two cache 112, a processor core 114, and registers 116. The example processor core 114 may include an Arithmetic Logic Unit (ALU), a Floating Point Unit (FPU), a digital signal processing core (DSP core), or any combination thereof. The example memory controller 118 may be used with the processor 104, or in some implementations the memory controller 118 may be an internal part of the processor 104.

Depending on the desired configuration, system memory 106 may be any type of memory, including but not limited to: volatile memory (such as RAM), non-volatile memory (such as ROM, flash memory, etc.), or any combination thereof. System memory 106 may include an operating system 120, one or more applications 122, and program data 124. In some embodiments, application 122 may be arranged to operate with program data 124 on an operating system. The program data 124 comprise instructions, and in the computing device 100 according to the invention the program data 124 comprise instructions for performing the method 200 for planning an electric vehicle parking lot.

Computing device 100 may also include an interface bus 140 that facilitates communication from various interface devices (e.g., output devices 142, peripheral interfaces 144, and communication devices 146) to the basic configuration 102 via the bus/interface controller 130. The example output device 142 includes a graphics processing unit 148 and an audio processing unit 150. They may be configured to facilitate communication with various external devices, such as a display or speakers, via one or more a/V ports 152. Example peripheral interfaces 144 may include a serial interface controller 154 and a parallel interface controller 156, which may be configured to facilitate communication with external devices such as input devices (e.g., keyboard, mouse, pen, voice input device, touch input device) or other peripherals (e.g., printer, scanner, etc.) via one or more I/O ports 158. An example communication device 146 may include a network controller 160, which may be arranged to facilitate communications with one or more other computing devices 162 over a network communication link via one or more communication ports 164.

A network communication link may be one example of a communication medium. Communication media may typically be embodied by computer readable instructions, data structures, program modules, and may include any information delivery media, such as carrier waves or other transport mechanisms, in a modulated data signal. A "modulated data signal" may be a signal that has one or more of its data set or its changes made in such a manner as to encode information in the signal. By way of non-limiting example, communication media may include wired media such as a wired network or private-wired network, and various wireless media such as acoustic, Radio Frequency (RF), microwave, Infrared (IR), or other wireless media. The term computer readable media as used herein may include both storage media and communication media.

Computing device 100 may be implemented as a server, such as a file server, a database server, an application server, a WEB server, etc., or as part of a small-form factor portable (or mobile) electronic device, such as a cellular telephone, a Personal Digital Assistant (PDA), a personal media player device, a wireless WEB-watch device, a personal headset device, an application specific device, or a hybrid device that include any of the above functions. Computing device 100 may also be implemented as a personal computer including both desktop and notebook computer configurations. In some embodiments, the computing device 100 is configured to perform a method 200 of planning an electric vehicle parking lot according to the invention.

It should be understood that in the free market environment, the electric vehicle user can freely decide whether to enter into a charging contract with the electric vehicle integrator according to the behavior habit and preference of the user. If the electric vehicle is willing to contract with the electric vehicle integrator and determine the characteristics of its charging requirements (e.g., daily vehicle usage time and desired SOC), the electric vehicle parking lot may take over and provide corresponding services thereto. During operation, the electric automobile user obtains the charging service of the electric automobile parking lot and pays the service fee for purchasing energy to the electric automobile parking lot according to the established contract. If an electric vehicle user chooses not to contract with an electric vehicle parking lot user, they tend to meet their charging needs in the conventional charging mode. In this case, it is assumed that the electric vehicle user plugs directly into and draws power from the grid upon completion of the destination trip, i.e. uncontrolled charging, and immediately unplugs the power supply upon reaching the required SOC. Therefore, in this scenario, the charging costs for electric vehicle users are only from their energy usage, which is directly calculated by the electric grid based on the retail price.

In practice, the electric vehicle user may always be hesitant to select whether and which electric vehicle parking lot to select, since participating in the program not only brings benefits to him, but may also incur some additional costs. For example, distance travel inconvenience costs or battery degradation costs due to operation in a controlled mode. Therefore, in order to attract the owner of the electric vehicle to join the electric vehicle parking lot charging plan, the electric vehicle aggregator may adopt an incentive strategy and provide an incentive contract so that the earnings obtained by its participants are always greater than those in the conventional charging case. In fact, since the incentive of the electric vehicle directly affects the economy of the operation of the electric vehicle parking lot, the arrangement of the design contract and the incentive policy must be optimized according to the planning decision to ensure the profitability of the electric vehicle parking lot project.

Therefore, the invention provides a novel electric automobile parking planning method. Fig. 2 shows a flow diagram of a method 200 for planning an electric vehicle parking lot, according to an embodiment of the invention, adapted to be executed in a computing device, such as the computing device 100. As shown in fig. 2, the method begins at step S210.

In step S210, a regret mechanism model is constructed according to the reflection-reaction paradigm, and an electric vehicle user model based on decision-dependent uncertainty is constructed according to the regret mechanism model, where the electric vehicle user model includes a decision-dependent utility function.

The invention mainly relates to a BEV (fully electric vehicle). when each electric vehicle is taken as an independent individual, the time for each electric vehicle to start charging is a random variable, and the BEV obeys a probability density function for determining the traveling habits and the vehicle using habits of people, and simultaneously the initial state-of-charge (SOC) of a single electric vehicle is also a random function related to the total travel after the last charging. In such a case, in order to determine the chargeable and dischargeable power of a single electric vehicle, it is necessary to obtain a probability distribution for each electric vehicle trip. The probability distribution function of the travel distance d of the electric vehicle follows similar normal distribution:

wherein d is the travel stroke of the electric vehicle; mu is the average value of travel journey, and is generally 16.1 km; σ is the standard deviation of travel trip, typically 9.3 km. Assuming that the SOC and the travel distance of the electric vehicle are linear, the SOC of the electric vehicle at the beginning of charging can be according to the SOC_i＝(1-d/d_R) X 100% is estimated, where SOC_iIs the initial SOC of the electric vehicle, d is the travel distance of the electric vehicle, obeys the above normal distribution, d_RThe maximum travel distance of an electric vehicle is typically 100 km. Assuming that all electric vehicles return to the electric vehicle parking lot for charging in one day, a new probability density function can be obtained as follows:

the charge start time and initial SOC of a single electric vehicle are simple and easy to describe, but in an active distribution network or in an electric vehicle parking lot, there are hundreds of electric vehicles, which if each is regarded as an individual to be charged independently would greatly increase the difficulty of the optimization problem and make it difficult to perform optimization control, so the present invention considers the electric parking lot as a unit and introduces the concept of a virtual power plant. Namely, an overall control layer is constructed between the electric automobile and the power grid, the control layer is similar to the existing microgrid and energy collection concepts, but is more like a virtual power plant, because the electric automobiles in the electric automobile parking lot can be controlled, so that the overall effect on the power grid is shown. The virtual power plant is used for enabling some units which are not related to electricity to serve as a whole to a power grid after integrated scheduling, the virtual power plant plays the same role as a traditional power plant, shows the characteristics of the power plant, and helps the optimal control of an active power distribution network from some aspects.

The single electric automobile has strong randomness, although description can be carried out through a probability density function, optimization control cannot be carried out, when the electric parking lot model is considered, only SOC constraints and charging and discharging power constraints of the electric parking lot can be seen from a power grid side, and in order to obtain the SOC constraints and the charging and discharging power constraints of the electric parking lot under 24 sections, a typical scene needs to be obtained through clustering by a statistical method or a protocol needs to be signed with a user. The time of each vehicle arriving at and leaving the parking lot, the SOC of each vehicle arriving at the parking lot and the travel of each vehicle need to be known from the user side, which is considered if the modeling is refined; and if only from the perspective of the electric automobile parking lot, the modeling and the constraint of the electric automobile parking lot can be determined only by knowing the number of departures and arrivals of the electric automobile parking lot and the expectation of the driving journey of the electric automobile in each time section.

Suppose that each vehicle can only return to a fixed electric car park in one day and divide the day into 24 periods where 1 refers to a time slice from one point to two points, and 24 refers to a time slice from the beginning of the morning zero to one point. Suppose A_tNumber of electric vehicles arriving at electric parking lot at t-time section, D_tNumber of electric vehicles leaving parking lot at t-time section, N_tNumber of electric vehicles that can participate in scheduling, N_t＝N_t-1+A_t-D_tt is 1,2, … 24. For each time section N_tIn the control strategy of (1), D must be prepared_t+1The vehicle is about to leave at the next moment and must meet D in time section t_t+1The vehicle meets the travel demand of the user, i.e. SOC_setIn the model, the SOC of each vehicle is assumed_setAll are 80%, and the SOC of other vehicles is not lower than the minimum requirement (assumed to be 20%) for battery loss, so:

0.2·(N_t-D_t+1)·SOC_max+D_t+1·SOC_set≤SOC_Nt+P_t·Δt≤N_tSOC_maxt＝1,2,…24 (3)

SOC_Nt＝SOC_Nt-1+P_t-1·Δt+SOC_At-SOC_Dtt＝1,2,…24 (4)

in the formula SOC_NtIs t time section N_tTotal energy, SOC of electric vehicle capable of participating in dispatching_AtIs t time section A_tTotal energy, SOC, of an electric vehicle arriving at an electric parking lot_DtIs t time section D_tTotal energy of electric vehicle arriving at electric parking lot, SOC_AtCan be obtained from statistical data, SOC_DtAnd setting data for the electric automobile parking lot. Assuming that the electric energy SOC consumed by the electric automobile in one day_constantEqual to the electric energy absorbed from the electric network in one day at the electric vehicle stop, i.e. in

Generally, a primary motivation for selecting customers to participate in a charging plan provided by an electric vehicle aggregator is to gain economic benefit. However, in actual practice, since the charging plans provided by the electric vehicle aggregators are independently formulated and operated by themselves, electric vehicle users hardly know how much reward they will get from participating in the project before contracting with the electric vehicle aggregators. Therefore, in order to achieve the above objective, the most likely strategy is to estimate the profitability of the charging plan provided by the electric vehicle aggregator by observing the gains it has achieved before, and to take this into account in future activities. Due to this learning ability, the participation of electric vehicle users in a charging plan provided by an electric vehicle aggregator tends to follow the "reflection-response" (RR) paradigm in the kinematics, which can be interpreted as: for each decision, the user can switch from his current action to another if the previous results demonstrate that these actions would produce more enjoyable results or higher rewards. At the same time, strategies with higher rewards are to some extent considered to have higher probability. To properly account for this effect, the present invention constructs a Regret Mechanism (RM) model to describe the uncertainty of the electric vehicle user in "reflect-react" mode.

To describe modeling of the regret mechanism, the present invention uses a finite setRepresents a potential charging station for a user of the electric vehicle, where Ω_BAndrepresenting candidate nodes and regular charging nodes in the overall system. During the operation simulation, each electric automobile user i E omega_IIt will be assumed that one is faced with the problem of selecting which charging station to select at specified intervals, where each interval corresponds to each contract term, which may be considered four months (120 days). According to the definition of the regret mechanism, for any interval y, adopting a strategyIs dependent on the regret value of the userThe following were used:

wherein,is a binary variable of whether the electric vehicle i is willing to go to the potential charging station bb, usually 1 is willing and 0 is unwilling. When the subscript of the z parameter is added with variables such as tau or y, the binary variable indicates whether the tau time or the y time interval goes to the charging station or not. As shown in the above equation, the repentance value of the electric vehicle user is defined as the value after all results have occurred before a certain period of time, if previously selected(situation which has been realized in practice, e.g.) Replaced with a different selection, resulting in increased profits. Wherein the profit of the electric vehicle passes the utility function W_i:R₊→ R, and multiplied by a constant 1/y to normalize each interval. Equation (5) shows if the individual is from a policyHigher than previously selected strategyThe profit of (1) the user will regret his own decision and quantify the regret as G_y，ss，i. Otherwise he will not have any regret, i.e. regret degree assignment:

based on the above definition, if an action is assumedIs taken, then at the y +1 thTime interval transition to new strategyThe probability of (c) can be described by the decision-dependent model as follows:

whereinRepresentation selection policyThe probability of (d); mu.s_iIs a predefined constant that ensures that the sum of all probabilities is 1. Equation (8) shows that for each operating period, the electric vehicle user can retain his previous policyOr from Ω_BBSelecting a new strategyThe probability of each strategy being selected is a value related to the regret value M_y，ss，i(.), the greater the regret value, the higher the probability that it is likely to be selected. Furthermore, if each electric vehicle user under study behaves according to the principles of equation (8), thenThe actual distribution of (c) will eventually converge to the correlated equilibrium set t → ∞. At this point of equilibrium, all electric vehicle users will not regret their chosen decisions, as everyone's decisions can take into account the others ' potential policies and translate into everyone's optimal system revenue. The regret model is completely consistent with the assumption of RR paradigm, and can well describe the situation that the electric vehicle user is excited by planning decision and contractThe behavior of (c).

According to equation (8), the probability of selecting a parking lot by the electric vehicle user will vary depending on the relative decision of the electric vehicle integrator, i.e.Wherein Rew_bbIs the incentive price for potential charging node bb,is a binary variable of whether the electric vehicle i has signed a contract with the potential charging station bb, typically 1 is signed, 0 is not signed,respectively charging power and discharging power at the potential charging site bb. This results inThe results of (c) act as endogenous (decision-dependent) uncertainties in our model. In actual optimization, given historical decisions of the electric vehicle integrator EVA, algorithm 1 can be used to determine the PEVi ∈ Ω of each plug-in electric vehicle in the interval y_IProbability distribution ofGeneratingThe method of probability distribution is as follows:

in a decision-dependent uncertainty model of the electric automobile, a utility function is used as an important index for characterizing each candidate scheme. According to one embodiment, the revenue W of an electric vehicle i participating in an electric vehicle parking lot charging plan_y，iConsidered as reward and subsidy for each interval yReduce the inconvenient costAnd cost of battery depletionThat is, the decision dependent utility function may be:

wherein,is the required state of charge SOC when the electric automobile i leaves the parking lot,is the state of charge of the electric vehicle when i arrives at the parking lot,andwhether the electric vehicle i is willing to go to the potential charging station bb or not at the time interval y, respectivelyThe binary variable signed up with the charging station bb, θ is the number of days covered by one time interval y,is the energy capacity (kWh) of the batteries of the electric automobile,andrespectively the departure time and arrival time, Rew, of the electric vehicle i_1,bbAnd Rew_2,bbRespectively the residence time of the potential charging node bb and the excitation price of the charging electric quantity, h_bb,iIs the equivalent distance (km) between the destination and the electric car parking lot,is the distance cost ($/km), π^dgIs the degradation cost ($/kWh) of the electric vehicle battery, T is a period of time T a day,is the battery depletion cost in the normal charging mode.

The conventional charging cost is generally deducted from equation (12)To ensureThe result of (a) reflects only the "incremental" cost of the electric vehicle user to participate in the electric vehicle parking lot dispatching. In the process of describing utility function, in order to calculateWe assume that electric vehicle integrators have adopted advanced equity-based scheduling strategies for real-time scheduled operation of batteries, and thus for electric vehicle users, electric vehiclesThe battery deterioration caused by the car park operation can always be regarded as the average distribution of all the electric car users participating in the operation in the current time period. Furthermore if the user of the electric vehicle selects the regular charging mode, i.e.His profit W_y，iWill be 0.

Subsequently, in step S220, an uncertainty scene model of the electric vehicle is constructed by clustering the uncertainty of the electric vehicle, the uncertainty scene model including all states that may occur in the electric vehicle parking lot during operation.

It should be appreciated that uncertainties associated with electric vehicles can generally be attributed to two aspects: selection of travel mode and charging scheme. The former uncertainty is exogenous in that their statistical characteristics are fixed and can be represented by a predetermined probabilistic model, while the latter uncertainty is endogenous in that its probability is not constant but evolves with the decision of the electric vehicle aggregator. According to one embodiment, the uncertainty of the electric vehicle comprises an endogenous uncertainty and an exogenous uncertainty, wherein the endogenous uncertainty comprises a willingness factor for participation of a user of the electric vehicleThe exogenous uncertainty includes the initial state of charge, arrival time, and departure time of the electric vehicle. That is, the information about the behavior of each electric vehicle user i can be represented by the vector Φ_iExpressed as:

through the composite statistics of all PEVs in the system, the modelable charging demand scenario can be represented as:

Ψ_s＝{Φ_i|i∈Ω_I} (14)

therein, Ψ_SIs a representation of a scene s, representing all possible states of the electric car park during operation, Ω_IIs a set of electric vehicle users. If suppose Φ_iAll uncertain variables in (a) are independent, then the operating scenario of the bus parking lot in each interval y can be generated by a Monte Carlo Simulation (MCS) process, as shown in algorithm 2:

subsequently, in step S230, a two-stage planning model of the electric vehicle parking lot is constructed according to the electric vehicle user model and the uncertainty scene model, and the two-stage planning model aims at obtaining the maximum profit of the electric vehicle integrator in the power distribution network, and comprises upper-layer planning of parking lot site selection, volume fixing, price incentive design and lower-layer planning of parking lot operation scheduling.

FIG. 4 illustrates a planning framework for a two-phase model according to one embodiment of the invention. The framework divides the decision field into two phases, the first phase mainly aims at processing position planning, capacity selection and contract (incentive) design related to the electric automobile parking lot, which corresponds to the decision of the electric automobile integrator in the planning phase; and in the second stage, evaluating the economy of the proposed solution by running a simulation program while considering various electric vehicle running scenarios. In order to realize effective simulation of the electric automobile, the exogenous and endogenous uncertainties of the electric automobile are considered in the framework. The internal uncertainty and the external uncertainty of the electric vehicle are generated and represented by a scene set, the internal uncertainty is that the participation willingness factor of the electric vehicle user is related to the planning decision and the operation decision of the electric vehicle through a regret degree mechanism, namely the decision depends on the uncertainty described before.

On the basis of carrying out statistical scene modeling on models of endogenous uncertainty and exogenous uncertainty of electric vehicle users and scenes, the method takes the maximum profit obtained by electric vehicle integrators in the power distribution network as an objective function, makes all planning decisions and excitation contract designs, and simultaneously enables the electric vehicles in the whole power grid to obtain the optimal output and the plan value of power absorption from the main network in an operation period. Because the decision of the aggregator of the electric automobile has great influence on the operation of the electric automobile, the optimal location and incentive policy design of the electric automobile parking lot is realized by adopting a two-stage random planning model.

According to one embodiment of the invention, the upper-level objective function of the two-phase planning model is profit F of the electric vehicle integrator in the power distribution network^PLMaximization, which is calculated by the formula:

max F^PL＝Λ^Ope-C^Inv(15)

wherein, Λ^OpeAnnual operating income, C^InvIs the annual equivalent investment cost,is the number of electric vehicle charging stations set at node b,is a binary variable, omega, representing whether the electric car park is established at node b_BIs a candidate node in the system, omega_SIs a set of scenes, Y is a contract period,ρ_y,sis the probability of the scene s occurring at the time interval y, Λ_y,sIs the operating revenue, k, of scene s at time interval y^cpIs a year-valued operator, k, of the charging pile^ldIs a annual value operator of the land, pi^cpIs the investment cost of the bidirectional charger.

In order to keep the investment cost consistent with the time scale of the operating cost, the invention converts the annual equivalent investment cost and annual operating income into the equal annual value k ═ ζ (1+ ζ) by using the capital recovery factor^d]/ [(1+ζ)^d-1]Where ζ represents the lifetime of the equipment and d represents the annual average discount rate.

According to one embodiment of the invention, the upper level constraints of the two-phase planning model are:

wherein,is the maximum number of electric vehicle charging stations, Rew, set at node b_bThe price of incentive subsidy is set at the node b, namely the incentive fee signed by the electric automobile parking lot and the electric automobile user,is the maximum incentive subsidy price set at node b.

Because of Λ^OpeCorresponding to the revenue expected from electric vehicle integrators based on the grid market and electric vehicle charging and discharging interactions. In fact, the criteria of fees levied by each electric vehicle parking plant depend on a bilateral agreement between the electric vehicle integrator and its customers, i.e. an incentive contract. Is not at oneIn the general case, it is assumed that the charging cost of the electric car parking lot is always equal to that of the conventional case, that is:

wherein,indicating the charging fee of the electric vehicle, β_y，s，iRepresents the total time that the battery of the electric vehicle is fully charged, theta represents the number of days in each operating cycle,the time of arrival is indicated by the time of arrival,indicating the nominal charging efficiency in the normal charging mode,the real-time electricity rate is represented,indicating the electric vehicle state of charge required at departure,indicating the state of charge of the electric vehicle at arrival,energy capacity of an electric vehicle battery, η^slIndicating the rated charging efficiency of the electric vehicle.

According to another embodiment of the invention, the lower-layer objective function of the two-stage planning model maximizes the operating profit of the parking lot, and the decision variables comprise the charge and discharge power of the electric automobile parking lot and the binary variables of whether the user signs a contract with the electric automobile parking lot. The lower-level objective function of the two-phase planning model is as follows:

wherein,represents the operation income of the electric automobile parking lot,represents the running cost of the electric car parking lot,represents the contract customization fee of the electric automobile parking lot,represents the discharge power of the electric car parking lot,represents the discharge electricity rate of the electric vehicle, r is an operation time interval,a binary variable indicating whether or not the electric vehicle user enters a price incentive contract with the s parking lot,represents the charging cost of the electric vehicle, pi^omRepresents the daily maintenance cost of the electric vehicle,represents the charging power of the electric car parking lot,the real-time electricity rate is represented,is a binary variable of whether the electric vehicle i has contracted a contract with the candidate parking lot b at the time interval y,respectively representing the departure time and arrival time of the electric vehicle i at the scene s of the time interval y,represents the state of charge of the electric vehicle when arriving,indicating the state of charge, Rew, that the electric vehicle needs to reach_1,bAnd Rew_2,bRespectively, the stay time of the candidate electric automobile parking lot b and the incentive electricity price of the charging electric quantity.

According to another embodiment of the invention, the lower layer constraint conditions of the two-stage planning model comprise at least one of maximum charge and discharge power constraint, constraint that charge and discharge cannot be simultaneously carried out, electric vehicle state of charge constraint, constraint that electric vehicle charging demand is met, electric vehicle battery loss constraint, electric vehicle available binary constraint, electric vehicle contract amount constraint, electric vehicle available amount constraint, electric vehicle arrival amount constraint and electric vehicle departure amount constraint. Wherein, the maximum charge and discharge power constraint is as follows:

wherein, γ^maxRepresents the maximum charge and discharge power of the charging pile of the electric automobile,represents the schedulable number of electric vehicles in the electric vehicle parking lot, T being a period of time of day T.

In addition, the constraint that charging and discharging can not be carried out simultaneously, the constraint of the SOC of the electric automobile and the constraint of the battery loss of the electric automobile are respectively represented by the following three formulas:

wherein E is_y,s,b,tRepresenting the total state of charge of the electric vehicle at the current stage, E_y,s,b,t-1Representing the total state of charge of the electric vehicle in the previous stage, η representing the charge-discharge efficiency,indicating the number of arriving electric vehicles in the electric vehicle parking lot,indicating the number of exiting electric vehicles in the electric vehicle parking lot. Pi^dgThe cost of the battery loss of the electric automobile is represented, and psi represents the limit constant of the battery loss of the parking lot of the electric automobile.

The available binary constraints for electric vehicles are:

wherein,a binary variable indicating whether the plug-in electric vehicle is in a plug-in state,andare binary variables of whether the electric vehicle arrives and departs, respectively.

Electric automobile available quantityConstrained electric vehicle arrival volumeRestraint and electric vehicle departureThe constraints are respectively the following three formulas:

subsequently, in step S240, the upper-layer plan and the lower-layer plan of the two-stage planning model are solved by using a predetermined algorithm, respectively, so as to obtain an optimal planning scheme for the electric vehicle parking lot. The upper-layer planning algorithm in the first stage may be a genetic algorithm, the lower-layer planning algorithm in the second stage may be a meta-dual interior point method, and specific parameter details of the algorithm may be set by a person skilled in the art according to actual needs, which is not limited by the present invention.

The present invention employs a two-stage stochastic programming solution method based on Genetic Algorithms (GA) that are used to account for first-stage variables in upper-level programming, e.g.And delta_b，Then, the traditional primal-dual interior point method (PIPM) is adopted to solve the operation optimization problem of the lower layer, and the profit of each operation period is estimated while the operation constraint of the electric automobile parking lot and the behavior constraint of the electric automobile are considered. Corresponding operating results (s∈Ω_S，b∈Ω_B，t∈T，i∈Ω_I) The fitness value will be returned to the decision of the first phase run plan. By repeating the iterative operation, the optimal solution of the entire planning operation model can be finally obtained based on the process in fig. 5. In the genetic algorithm-based algorithm, candidate solutions for electric vehicle integrators are represented by a series of randomly created chromosomes. Each population member has a total of 3 × Ω_BThe constituent parts (genes),indicating the location of the electric car parking lot configuration,indicating the number of charging points to be installed, Rew_bA value representing an incentive contract provided for each electric vehicle parking place. Evaluating the performance of each chromosome by using a fitness function to compare the goodness and badness of each chromosome, and obtaining a suggested planning decision: fitness ═ F^PL-PF Fitness＝F^PL-PF, wherein F^PLDenotes the value OF defined by this equation, PF is a penalty factor.

If chromosome individuals violate constraints in the simulation, i.e., strictly speaking, the underlying optimization does not converge to the second-stage optimization, the penalty factor will be set to a large number (10)⁸) Unlike traditional genetic algorithms, the modified method maintains individual diversity by using an adaptive selection operator to generate populations by minimizing the similarity of niche clusters defined by spatial distance_max) Or maximum iteration time without adaptive improvement (β)_uch)。After the algorithm is terminated, the best individual in the whole population is considered as the final solution of the whole electric vehicle parking lot planning operation model.

The invention provides a charging facility planning and incentive contract customizing framework of an electric automobile parking lot based on decision-dependent uncertainty for promoting effective integration of electric automobiles under the condition of a power distribution system accessed by high-permeability electric automobiles, establishes a two-stage optimization model, describes dynamic distribution of participation rate of electric automobile users and decision-dependent action of electric automobile integrators by utilizing a regret degree mechanism, and repeatedly researches an electric automobile parking lot planning method based on decision-dependent uncertainty.

Fig. 3 shows an electric car park planning apparatus 300 according to an embodiment of the present invention, which is adapted to be resident in a computing device for execution. As shown in fig. 3, the apparatus includes a user model building unit 310, a scene model building unit 320, a planning model building unit 330, and a planning model solving unit 340.

The user model construction unit 310 is adapted to construct a regret mechanism model according to the reflection-reaction paradigm, and construct an electric vehicle user model based on decision-dependent uncertainty according to the regret mechanism model, where the electric vehicle user model includes a decision-dependent utility function.

The scenario model construction unit 320 is adapted to construct an uncertainty scenario model of the electric vehicle by clustering the uncertainty of the electric vehicle, the uncertainty scenario model comprising all states that may occur in the electric vehicle parking lot during operation.

The planning model construction unit 330 is adapted to construct a two-stage planning model of the electric vehicle parking lot according to the electric vehicle user model and the uncertainty scene model, wherein the two-stage planning model aims at obtaining the maximum profit of an electric vehicle integrator in the power distribution network, and comprises an upper-layer planning of parking lot site selection, volume fixing and price incentive design and a lower-layer planning of parking lot operation scheduling.

The planning model solving unit 340 is adapted to respectively adopt a predetermined algorithm to solve the upper-layer plan and the lower-layer plan of the two-stage planning model, so as to obtain an optimal planning scheme of the electric vehicle parking lot.

The details of the planning apparatus 300 for an electric car park according to the present invention are disclosed in detail in the description based on fig. 1-5, and are not described herein again.

In the following, the model proposed by the present invention is calculated and planned by using an improved IEEE12 node system (as shown in fig. 6), and knowing parameters such as IEEE12 system deployment, rated voltage, etc., and parameters such as electric vehicle capacity, maximum driving mileage, time for fast charging to 80% electric quantity, etc., human eyes in the art can obtain P according to the prior art^slInconvenient cost pi^duThe system comprises a plurality of electric automobile charging piles.

To verify the validity of the proposed framework, five different schemes were set up to simulate in the case of an incentive scheme that only considers individual price subsidies. In the first scheme, the exogenous uncertainty of the electric automobile is considered, and an incentive policy is adopted to incentivize users of the electric automobile, so that the electric automobile is signed with a parking lot of the electric automobile; in other schemes, the exogenous uncertainty and the endogenous uncertainty of the electric automobile are considered at the same time, namely, the electric automobile user is considered to react according to an incentive policy adopted by the parking lot, so that the planning and the operation of the whole electric automobile operator and the parking lot of the electric automobile are influenced. The set scene specific information is shown in table 1. The scheme I and the scheme II adopt lower incentive amounts, the scheme III adopts relatively higher incentive amounts with higher cost, the scheme IV sets the incentive amounts as an optimization variable in an upper-layer genetic algorithm, the algorithm provided by the paper is used for carrying out optimization solution, and the scheme V sets the incentive amounts as an optimization variable related to the electric automobile station in the upper-layer genetic algorithm, namely the incentive amounts of each electric automobile parking lot obtained by optimization solution are different, so that the planning method and the incentive policy mechanism provided by the whole model are analyzed and researched more comprehensively.

TABLE 1 exemplary scenarios

The two-stage stochastic programming solving algorithm constructed by the invention can be used for solving all the example scenes in the table 1, and the table 2 gives the location and volume optimization results in the upper-layer programming of the algorithm and the optimal amount of money of the incentive policy of the electric automobile parking lot. The overall upper layer planning result can be described as follows: the method comprises the steps that 50 charging piles are respectively built on nodes 633, 645 and 675 in scheme 1, 50 charging piles are built on nodes 633 in scheme 2, 38 charging piles are built on nodes 633 and 44 charging piles are built on nodes 645 in scheme 3, 43 charging piles are built on nodes 633 and 50 charging piles are built on nodes 645, incentive policy price is set to be $ 79.72 per period, 50 charging piles are built on nodes 633 in scheme 5, incentive policy price is set to be $ 91.63 per period, 50 charging piles are built on nodes 645, and incentive policy price is set to be $ 74.72 per period.

TABLE 2 upper layer planning results

Meanwhile, in order to have more deep knowledge on the whole two-stage model and better analyze and evaluate the whole planning result, the invention lists the EVA cost/profit response values calculated by the lower-layer operation result under each scene in the table 3:

table 3 run phase results (dollar $)

Specific construction investment cost, maintenance cost, incentive cost, operation income and net income of each scheme are listed specifically, so that the difference of each scheme can be analyzed and compared conveniently, and specific reasons influencing the planning result can be analyzed. The focus of the present invention is to exploit the organic combination of decision dependent uncertainties (endogenous uncertainties) and PEVs in electric vehicle parking lot planning and incentive customization decisions. The electric car parking lots under each scheme can be analyzed by studying the change of expected Participation Rate (PR) of each user of each electric car parking lot in each excitation periodWherein the expected participation rate of the electric vehicle parking lot b in the scene s is defined as the number of electric vehicles willing to contract with the electric vehicle parking lot in the interval y, i.e. the electric vehicle selection

In the first scheme, the expected participation rate of the electric automobile parking lot seems to be stable over time, and a larger expected participation rate means that the utilization efficiency of the electric automobile parking lot is higher, and an electric automobile integrator can obtain a larger return on investment during operation. However, when including the effect of endogenous uncertainty, i.e. there is a decision-dependent uncertainty, the charging behaviour of the electric vehicle user is associated with the decision of the electric vehicle integrator. In the second scheme, the influence of the incentive is very low, the expected participation rate of the electric automobile parking lot will be reduced sharply with the passage of time, and the final profit of the electric automobile integrator is far lower than that of the first scheme under the same planning decision result. To illustrate the advantages of the decision-dependent approach, the value of the endogenous uncertainty (VEU) is quantified by inserting the solution of scheme 1 into the application scenario of scheme two. The target value for solution one under the decision-dependent model is calculated as $ 5370.26. Where a positive value of VEU ($ 10289.03- $ 5370.26 $ 4918.77) indicates that the solution obtained in solution two can provide a greater economic benefit to the electric vehicle integrator than that obtained in solution one. Therefore, the framework proposed by the invention has better results in practical application than the traditional random planning method.

In practice, because individuals may have different behavior patterns according to different behavior characteristics, inherent uncertainties of electric vehicle users, that is, decision-dependent uncertainties of electric vehicle users for electric vehicle parking lot planning decisions and incentive decisions, should be explicitly considered in the decision-making of electric vehicle parking lot planning. Otherwise, the electric vehicle integrator may overestimate the profitability of the electric vehicle parking lot planning project and make ineffective investment decisions. When the incentive compensation provided by PL changes, as in plan three Rew for electric car park_bAt higher times, the electric vehicle user expects less change in engagement rate over time, which means that the electric vehicle parking lot can be more efficiently utilized during operation. However, the greater incentive fee does not guarantee the electric vehicle integrator a higher profit. When Rew_bUp to $ 150 per dispatch period, the total revenue for the electric vehicle integrator is far lower than for scenario two. That is, with the cost of excitation, Rew_bThe income of electric vehicle integrators tends to increase, but the operating cost thereof also rises. However, the profit of the electric vehicle integrator is generally lower than that of solution two, since the increase in revenue is not as great as the increase in cost.

Therefore, in order to maximize the benefits of electric vehicle integrators, it is crucial to design incentive contracts that simultaneously consider electric vehicle users within the same framework. And in the fourth scheme, a fixed incentive policy is not adopted any more, and the incentive policy is used as a decision variable in a planning stage, so that the participation rate of the electric automobile is improved under the optimization condition. And in the fifth scheme, the incentive policy is optimized by adopting different incentive methods of different nodes, so that the profit maximization of the electric vehicle aggregator is realized. As shown in table 3, the optimal incentive measures may improve the profits of the electric vehicle aggregators of scenarios four and five. Moreover, better excitation effect can be obtained by adopting the node-based excitation scheme, thereby proving the advantages of the comprehensive optimization framework constructed in the invention.

In the above analysis, it is assumed that incentive rewards paid to users contracting with electric vehicle aggregators have been implemented through a fixed-based scheme. The model is then simulated (as defined in equation (10)) using different excitation structures to determine the variation in its simulation results. The electric quantity excitation (shown in formula (10B)) is subsidized according to the charge quantity of the electric vehicle, and the comprehensive excitation (shown in formula (10B)) is subsidized according to the charge quantity of the electric vehicle and the time for staying in the parking lot.

Table 4 run phase results (dollar $)

It can be seen that the plan under the integrated incentive conditions provides greater profit on the electric vehicle aggregation than under the fixed incentive and electric incentive schemes (B)^IPL$ 40093.13). In a fixed incentive scheme, electric car aggregators must pay fixed rewards to users contracting with electric car parks to reward them for participating in the dispatch run. Thus, building an electric car park with a firm customer base would require a significant expenditure for the contract of incentive contracts, but with a low return efficiency, the planning solution under the fixed incentive scheme seems to be economically inefficient. Whereas under the electric quantity incentive scheme, the remuneration received by the electric vehicles depends on the energy (charging capacity) they draw from the electric vehicle parking lot ). The higher the charge demand, the more revenue is obtained. Therefore, owners of electric vehicles with short travel distances per day can only obtain very limited revenue if they participate in the contractual incentive program, thereby reducing the chances of using this option for charging during operation. This negative effect will undoubtedly reduce the investment efficiency. However, in the solution of the integrated incentive, the electric vehicle user is rewarded not only in terms of energy consumption but also in terms of scheduling availability of the electric vehicle parking lot. Since the availability payment can fairly reflect the actual contribution of the electric vehicle user to the system, the method provides stronger power for the customer (especially short trip user) to join the contract customization scheme of the electric vehicle parking lot, thereby bringing greater benefits for the electric vehicle integrator.

According to the technical scheme of the invention, on the basis of an endogenous uncertainty and exogenous uncertainty model of an electric automobile user and a scene statistical scene model, all planning decisions and excitation contract designs are made by taking the maximum profit obtained by an electric automobile integrator in a power distribution network as an objective function, and meanwhile, the electric automobiles in the whole power grid obtain the optimal output and the plan value of power absorption from a main network in an operation period. Meanwhile, a two-stage random planning model is adopted to realize the optimal site selection and incentive policy design of the electric automobile parking lot, the planning model of the electric automobile parking lot based on decision-dependent uncertainty is established, and a two-stage solving algorithm is adopted to solve the model to obtain the optimal planning of the electric automobile parking lot.

A8, the method as in A7, wherein the lower level objective function of the two-phase planning model is: wherein,represents the operation income of the electric automobile parking lot,represents the running cost of the electric car parking lot,represents the contract customization cost of the electric automobile parking lot, theta represents the number of days in each operation period, omega_IIs a set of users that are in a group,represents the discharge power of the electric car parking lot,represents the discharge electricity rate of the electric vehicle, r is an operation time interval,a binary variable indicating whether or not the electric vehicle user enters a price incentive contract with the s parking lot,represents the charging cost of the electric vehicle, pi^omRepresents the daily maintenance cost of the electric vehicle,represents the charging power of the electric car parking lot,indicating the real-time electricity price, Rew_bRepresenting the incentive fee signed by the electric automobile parking lot and the electric automobile user.

The method A9 is as described in A7, wherein the lower layer constraint conditions of the two-stage planning model include at least one of maximum charge and discharge power constraint, constraint that charge and discharge cannot be performed simultaneously, electric vehicle state of charge constraint, constraint that electric vehicle charging demand is met, electric vehicle battery loss constraint, electric vehicle available binary constraint, electric vehicle contract amount constraint, electric vehicle available amount constraint, electric vehicle arrival amount constraint and electric vehicle departure amount constraint. A10, the method of claim 9, wherein the maximum charge-discharge power constraint is:

wherein, γ^maxRepresents the maximum charge and discharge power of the charging pile of the electric automobile,represents the schedulable number of electric vehicles in the electric vehicle parking lot, T being a period of time of day T. A11, the method of a9, wherein the constraint that charging and discharging cannot be performed simultaneously is:

a12, the method of A9, wherein the electric vehicle state of charge constraints are:

A13, the method of A9, wherein the electric vehicle battery loss constraint is:

wherein, pi^dgRepresents the battery loss cost of the electric automobile, psi represents the battery loss limit constant of the electric automobile parking lot,representing the nominal charging power in the normal mode.

The various techniques described herein may be implemented in connection with hardware or software or, alternatively, with a combination of both. Thus, the methods and apparatus of the present invention, or certain aspects or portions thereof, may take the form of program code (i.e., instructions) embodied in tangible media, such as floppy diskettes, CD-ROMs, hard drives, or any other machine-readable storage medium, wherein, when the program is loaded into and executed by a machine, such as a computer, the machine becomes an apparatus for practicing the invention. In the case of program code execution on programmable computers, the computing device will generally include a processor, a storage medium readable by the processor (including volatile and non-volatile memory and/or storage elements), at least one input device, and at least one output device. Wherein the memory is configured to store program code; the processor is configured to execute the method for planning an electric vehicle parking lot according to the invention, according to the instructions in said program code stored in the memory. By way of example, and not limitation, computer readable media may comprise computer storage media and communication media. Computer-readable media includes both computer storage media and communication media. Computer storage media store information such as computer readable instructions, data structures, program modules or other data. Communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media. Combinations of any of the above are also included within the scope of computer readable media.

In the description of the invention, the algorithms and displays are not inherently related to any particular computer, virtual machine, or other apparatus. Various general purpose systems may also be used with examples of this invention. The required structure for constructing such a system will be apparent from the description above. Moreover, the present invention is not directed to any particular programming language. It is appreciated that a variety of programming languages may be used to implement the teachings of the present invention as described herein, and any descriptions of specific languages are provided above to disclose the best mode of the invention.

Those skilled in the art will appreciate that the modules or units or components of the devices in the examples disclosed herein may be arranged in a device as described in this embodiment or alternatively may be located in one or more devices different from the devices in this example. The modules in the foregoing examples may be combined into one module or may be further divided into multiple sub-modules. Those skilled in the art will appreciate that the modules in the device in an embodiment may be adaptively changed and disposed in one or more devices different from the embodiment. The modules or units or components of the embodiments may be combined into one module or unit or component, and furthermore they may be divided into a plurality of sub-modules or sub-units or sub-components. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where at least some of such features and/or processes or elements are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise. Furthermore, some of the described embodiments are described herein as a method or combination of method elements that can be performed by a processor of a computer system or by other means of performing the described functions. A processor having the necessary instructions for carrying out the method or method elements thus forms a means for carrying out the method or method elements.

Claims

1. A method of planning an electric vehicle parking lot, adapted to be executed in a computing device, the method comprising:

constructing a repentance mechanism model according to a reflection-reaction paradigm, and constructing an electric vehicle user model based on decision-dependent uncertainty according to the repentance mechanism model, wherein the electric vehicle user model comprises a decision-dependent utility function;

constructing an uncertainty scene model of the electric vehicle by performing cluster analysis on uncertainty of the electric vehicle, wherein the uncertainty scene model comprises all states of the electric vehicle parking lot which can occur during operation;

constructing a two-stage planning model of the electric automobile parking lot according to the electric automobile user model and the uncertainty scene model, wherein the two-stage planning model aims at obtaining the maximum profit of an electric automobile integrator in the power distribution network and comprises upper-layer planning of parking lot site selection, volume fixing and price incentive design and lower-layer planning of parking lot operation scheduling; and

and respectively solving the upper-layer plan and the lower-layer plan of the two-stage planning model by adopting a predetermined algorithm to obtain an optimal planning scheme of the electric automobile parking lot.

2. The method of claim 1, wherein the decision dependent utility function comprises:

3. The method of claim 1, wherein the uncertainty of the electric vehicle comprises an endogenous uncertainty and an exogenous uncertainty, wherein the endogenous uncertainty comprises a willingness factor for participation of a user of the electric vehicle, and the exogenous uncertainty comprises an initial state of charge, an arrival time, and a departure time of the electric vehicle.

4. The method of claim 1, wherein the upper-level planning algorithm of the first phase is a genetic algorithm and the lower-level planning algorithm of the second phase is a meta-dual interior point method.

5. The method of any of claims 1-4, wherein the upper-level objective function of the two-phase planning model is profit F of an electric vehicle integrator in the power distribution grid^PLMaximization, which is calculated by the formula:

max F^PL＝Λ^Ope-C^Inv

wherein, Λ^OpeAnnual operating income, C^InvIs the annual equivalent investment cost,is the number of electric vehicle charging stations set at node b,is a binary variable, omega, representing whether the electric car park is established at node b_BIs a candidate node in the system, omega_SIs a scene set, Y is a contract period, ρ_y,sIs the probability of the scene s occurring at the time interval y, Λ_y,sIs the operating revenue, k, of scene s at time interval y^cpIs a year-valued operator, k, of the charging pile^ldIs a annual value operator of the land, pi^cpIs the investment cost of the bidirectional charger.

6. The method of claim 5, wherein the upper level constraints of the two-phase planning model are:

7. The method of any one of claims 1-6, wherein a lower layer objective function of the two-phase planning model maximizes a parking lot operating yield, and the decision variables include charge and discharge power of an electric car parking lot and binary variables of whether a user signs up with the electric car parking lot.

8. An apparatus for planning an electric car park, adapted to be resident in a computing device for execution, the apparatus comprising:

the user model building unit is suitable for building a regret degree mechanism model according to a reflection-reaction paradigm, and building an electric vehicle user model based on decision-dependent uncertainty according to the regret degree mechanism model, wherein the electric vehicle user model comprises a decision-dependent utility function;

the scene model building unit is suitable for building an uncertainty scene model of the electric automobile through clustering analysis of uncertainty of the electric automobile, wherein the uncertainty scene model comprises all states of the electric automobile parking lot which can occur during operation;

the planning model construction unit is suitable for constructing a two-stage planning model of the electric automobile parking lot according to the electric automobile user model and the uncertainty scene model, the two-stage planning model aims at obtaining the maximum profit of an electric automobile integrator in the power distribution network and comprises upper-layer planning of parking lot site selection, constant volume and price incentive design and lower-layer planning of parking lot operation scheduling; and

and the planning model solving unit is suitable for respectively adopting a preset algorithm to solve the upper-layer planning and the lower-layer planning of the two-stage planning model so as to obtain the optimal planning scheme of the electric automobile parking lot.

9. A computing device, comprising:

at least one processor; and

at least one memory including computer program instructions;

the at least one memory and the computer program instructions are configured to, with the at least one processor, cause the computing device to perform the method of any of claims 1-7.

10. A computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by a server, cause the server to perform any of the methods of claims 1-7.