Electric vehicle cluster collaborative scheduling method for multi-scene vehicle network interaction

Abstract

The invention discloses an electric vehicle cluster collaborative scheduling method for multi-scene vehicle network interaction, and relates to the technical field of electric power system operation and control. The method comprises the steps that differentiated adjustable capacity modeling is conducted on five types of typical electric vehicle charging scenes, and the maximum charging and discharging power and the maximum adjustable energy of the electric vehicle charging scenes are accurately measured and calculated; constraint is constructed, gatherer income maximization is taken as a target, and a two-stage optimization model including day-ahead and real-time markets is established; the uncertainty of the bid winning probability of the climbing market is quantified by adopting an information gap decision theory, the model is converted into a robust optimization model, and the risk tolerance capability is maximized on the premise of ensuring the reference income; according to the method, the robust model is converted into a mixed integer linear programming problem to be solved, a bidding decision is obtained, finally, charging and discharging instructions of all scenes are generated and issued, the robust decision problem under heterogeneous electric vehicle cluster accurate evaluation and market uncertainty is solved, and the feasibility and safety of a scheduling scheme are guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system operation and control, in particular to a kind of electric vehicle cluster coordinated scheduling method for multi-scenario vehicle-grid interaction. BACKGROUND

[0002] High proportion of renewable energy represented by wind power and photovoltaic is widely accessed to power system, which brings significant fluctuation to power system, especially the rapid climbing event of net load poses a serious challenge to system real-time balance. Traditional generation side regulation resources have been difficult to meet the economic and efficient climbing demand, and the climbing auxiliary service market as a new auxiliary service mechanism aims to call flexible resources to provide climbing capacity.

[0003] The popularity of electric vehicles (EV) at scale makes its power battery a potential distributed energy storage resource, providing flexible regulation capacity for power system. Guiding EV to participate in the climbing auxiliary service market not only can provide the system with the much-needed fast and flexible climbing capacity, enhancing the safety of power grid operation, but also can provide a new business operation mode for vehicle-grid interaction (V2G), promoting the sustainable development of V2G market participation.

[0004] However, most of the existing researches use a single probability distribution (such as normal distribution or uniform distribution) to model different scenarios of electric vehicle cluster, which cannot accurately reflect the unique rules of each scenario, resulting in a large deviation between the calculated maximum charging and discharging power and the actual situation, and further affecting the effectiveness of the subsequent optimization scheduling scheme. And when participating in multi-time scale market transactions, EV aggregators need to deal with the dual influence of real-time market price and climbing service bid probability uncertainty. The existing scheduling strategy seriously depends on the accurate probability distribution assumption of uncertain parameters, which is difficult to obtain in actual application, resulting in unreliable optimization decision. SUMMARY

[0005] The technical problem to be solved by the present application is how to regulate and control the distributed electric vehicle cluster with high uncertainty and complex physical constraints to ensure its feasibility and safety when participating in power grid regulation. The purpose is to provide a kind of electric vehicle cluster coordinated scheduling method for multi-scenario vehicle-grid interaction, which solves the above problems.

[0006] The present application is realized by the following technical scheme: In a first aspect, the present application provides a kind of electric vehicle cluster coordinated scheduling method for multi-scenario vehicle-grid interaction, comprising: The adjustable energy of a plurality of typical electric vehicle charging scenes is respectively modeled, and the maximum charging and discharging power and the maximum adjustable energy of each scene are calculated; the plurality of scenes include a residential area household scene, a centralized commercial charging scene, a bus fleet scene, a city distribution logistics scene, and a heavy truck battery swap scene; Constraint conditions are constructed according to the maximum charging and discharging power and the maximum adjustable energy of each scene, and a two-stage optimization model including a day-ahead market and a real-time market is constructed, with the goal of maximizing the comprehensive income of an electric vehicle aggregator in the electricity market and the auxiliary service market; The two-stage stochastic optimization model is converted into an IGDT robust model by using information gap decision theory (IGDT) to quantify the uncertainty of the winning probability of the ramping market capacity; The IGDT robust model is converted into a mixed integer linear programming problem for solving, and the bidding decision that can still guarantee the minimum profit under the most unfavorable uncertainty is obtained; According to the bidding decision, charging and discharging power plan instructions for each scene are generated and issued to the charging facilities of each scene for execution.

[0007] Optionally, the adjustable energy of a plurality of typical electric vehicle charging scenes is respectively modeled, and the maximum charging and discharging power and the maximum adjustable energy of each scene are calculated, including: For the residential area household scene, the access time and the departure time of each resident vehicle are simulated by using a beta distribution, and the on-site state of each resident vehicle at each time is defined; The initial state of charge of each resident vehicle is simulated by using a truncated normal distribution; Based on the on-site state, the initial state of charge of each resident vehicle, the battery capacity parameter, and the charging facility capacity parameter, the maximum adjustable energy and the maximum charging and discharging power are calculated, and the formula is as follows:

[0008] In the formula, Q R,t is the maximum adjustable energy of the residential area household scene; P R,t is the maximum charging and discharging power of the residential area household scene; N R is the number of vehicles in the residential area; ξ R,i is the initial state of charge of the i-th resident vehicle; is the lower limit of the electric quantity of the resident vehicle; C R,i is the battery capacity of the i-th resident vehicle; indicates the on-site state of the i-th resident vehicle at time t ; is the maximum charging and discharging power of the i-th resident vehicle; is the number of available charging piles in the residential area; represents the maximum charging and discharging power of the cell; min represents the minimum value function.

[0009] Optionally, the adjustable capability modeling for various typical electric vehicle charging scenarios, and the calculation of the maximum charging / discharging power and maximum adjustable energy for each scenario, includes: For the centralized commercial charging scenario, a Poisson distribution is used to simulate the number of arriving vehicles at each time step; The initial state of charge of each commercial charging vehicle is simulated using a triangular distribution. Define the state variables for each commercial charging vehicle connected to the charging pile; Based on the number of arriving vehicles, the initial state of charge of each commercial charging vehicle, the state variables, battery capacity parameters, and charging facility capacity parameters, the maximum adjustable energy and maximum charging / discharging power are calculated using the following formulas:

[0010] In the formula, Q M,t This represents the maximum adjustable energy for the centralized commercial charging scenario. P M,t The maximum charging and discharging power for the aforementioned centralized commercial charging scenario; κ M,t For a moment t The number of vehicles arriving at centralized charging stations; ω M,j For the first j The initial state of charge of the commercial charging vehicle; The lower limit of charging power for commercial vehicles; C M,j The battery capacity of the j-th commercial vehicle; Ψ M,j,t For the first j State variables of commercial vehicles connected to charging piles; For the first j The maximum power of the commercial charging vehicle; The number of available charging piles at the centralized charging station; represents the maximum charging and discharging power within the centralized charging station; min represents the minimum value function.

[0011] Optionally, the adjustable capability modeling for various typical electric vehicle charging scenarios, and the calculation of the maximum charging / discharging power and maximum adjustable energy for each scenario, includes: For the bus fleet scenario, a Poisson distribution is used to simulate the number of buses at each time point. The station status variable for each bus is defined based on a fixed entry and exit timetable; The initial state of charge of each bus was simulated using a truncated normal distribution; Based on the number of buses, the on-site state variables, the initial state of charge of each bus, battery capacity parameters, and charging facility capacity parameters, the maximum adjustable energy and maximum charging / discharging power are calculated using the following formulas:

[0012] In the formula, Q B,t This represents the maximum adjustable energy for the bus fleet scenario. P B,t The maximum charging and discharging power for the bus fleet scenario; ν B,t The number of buses at station t; ζ B,k Let K be the initial state of charge of the k-th bus. This is the minimum battery level for buses; q B,k Y represents the battery capacity of the kth bus; B,k,t Let be the station-state variable of the k-th bus at time t; The maximum charging and discharging power of the k-th bus; The number of charging parking spaces at bus stops; The maximum charging and discharging power of the bus station is denoted by ; min represents the minimum value function.

[0013] Optionally, the adjustable capability modeling for various typical electric vehicle charging scenarios, and the calculation of the maximum charging / discharging power and maximum adjustable energy for each scenario, includes: For the aforementioned urban distribution logistics scenario, a Poisson distribution is used to simulate the number of delivery batches arriving per unit time, and a log-normal distribution is used to simulate the number of logistics vehicles arriving in each batch. The initial state of charge of each logistics vehicle is simulated using a uniform distribution. Based on the number of logistics vehicles, the initial state of charge of each vehicle, battery capacity parameters, and charging facility capacity parameters, the maximum adjustable energy and maximum charge / discharge power are calculated using the following formulas:

[0014] In the formula, Q L,t This represents the maximum adjustable energy for the urban distribution logistics scenario. P L,t This refers to the maximum charging and discharging power in the urban distribution logistics scenario. For the first m The initial state of charge of the logistics vehicle; η L,t For a moment t The number of logistics vehicles arriving at the station in each batch; This is the minimum battery capacity for logistics vehicles; Φ L,m,t For the firstm The state variables of a logistics vehicle; C L,m For the first m Battery capacity of a logistics vehicle; For the first m The maximum charging and discharging power of the logistics vehicle; The number of charging stations at the logistics station; The maximum charging and discharging power of the logistics station is denoted by ; min represents the minimum value function.

[0015] Optionally, the adjustable capability modeling for various typical electric vehicle charging scenarios, and the calculation of the maximum charging / discharging power and maximum adjustable energy for each scenario, includes: For the heavy-duty truck battery swapping scenario, the expected number of available batteries is calculated based on a queuing theory model; The initial state of charge of each backup battery is simulated using a beta distribution. Based on the expected number of available batteries, the initial state of charge of each backup battery, battery capacity parameters, and charging facility capacity parameters, the adjustable energy and maximum charge / discharge power are calculated using the following formulas:

[0016] where Q H,t This refers to the maximum adjustable energy for the heavy-duty truck battery swapping scenario. P H,t This refers to the maximum charging and discharging power in the heavy-duty truck battery swapping scenario. For a moment t The expected number of available batteries; This represents the initial state of charge of the nth battery. This is the lower limit of the battery capacity for heavy-duty truck battery swapping stations; C H,n Let n be the capacity of the nth battery. This refers to the number of battery swapping locations; This is the maximum power of the heavy-duty truck battery swapping station; denoted as the maximum charge / discharge power of the nth battery; min represents the minimum value function.

[0017] Optionally, the function of the comprehensive return is as follows:

[0018] Where max represents the maximization function; Q represents the overall return; This indicates the revenue from electricity generated by participating in the day-ahead market; This indicates the revenue from participating in the real-time market for electrical energy; This indicates the returns from participating in the uphill market; Indicates the returns from participating in the downhill market; C degThis represents the cost of battery degradation caused by charging and discharging operations in electric vehicles.

[0019] Optionally, the constraints include power constraints, regulation capacity constraints, and energy dynamic constraints; The power constraints are as follows:

[0020] The adjustment capacity constraint is as follows:

[0021] The energy dynamic constraints are as follows:

[0022] In the formula, This represents the day-ahead market charging level at time t in scenario s; This represents the day-ahead market discharge volume at time t in scenario s; This represents the application volume of the climbing market per unit time at time t in scenario s; express s In the scene t The application volume of the downhill market per unit time; P s,t,max For scene s t Maximum charging and discharging power at any given moment; express s In the scene t Real-time market charging power; express s In the scene t Real-time market discharge volume at any given moment; This indicates the minimum application capacity for the climbing market; This indicates the minimum application capacity for the downhill market; η cha Indicates charging efficiency; η dis Indicates discharge efficiency; Δt represents unit time; SOC min E represents the lowest initial state of charge; s express t Battery energy at any moment; E s,t,max express s In the scene t The maximum adjustable energy at any given time.

[0023] Optionally, the step of using Information Gap Decision Theory (IGDT) to quantify the uncertainty of the winning probability of the market capacity ramp-up, and transforming the two-stage stochastic optimization model into an IGDT robust model, includes: Define an uncertainty radius; the uncertainty radius includes a first uncertainty radius parameter and a second uncertainty radius parameter; the first uncertainty radius parameter represents the relative deviation of the probability of winning a bid in an upward climbing market; the second uncertainty radius parameter represents the relative deviation of the probability of winning a bid in a downward climbing market; Set the first and second uncertain radius parameters to zero, and when the winning probability in the climbing market is the benchmark value, solve the two-stage stochastic optimization model to obtain the benchmark profit; Based on the aforementioned benchmark profit, an acceptable minimum profit is set; With the goal of maximizing the tolerance for uncertainty and the constraint that the overall return is not less than the minimum profit, an IGDT robust model is constructed.

[0024] Optionally, the robust model of the IGDT is as follows:

[0025] Where max represents the maximization function; α 1 represents the first uncertain radius parameter; α 2 represents the second uncertain radius parameter; This indicates the probability of winning a bid in the uphill market; This indicates the probability of winning a bid in the downhill market. γ This represents the profit loss rate; z This represents the benchmark profit; Q This represents the overall benefit; x This represents the set of decision variables in the current market. y This represents the set of decision variables in a real-time market.

[0026] Compared with the prior art, the present invention has the following advantages and beneficial effects: This application provides a collaborative scheduling method for electric vehicle clusters in multiple scenarios involving vehicle-to-grid interaction. For five typical scenarios—residential communities, centralized commercial charging, bus fleets, urban logistics, and heavy-duty truck battery swapping—a probability distribution best matching their operational patterns is used for modeling. This multi-scenario differentiated modeling significantly improves the accuracy of adjustability assessment, ensuring the physical feasibility of the scheduling scheme from the outset. A two-stage optimization model—day-ahead and real-time—is constructed with the goal of maximizing the aggregator's overall revenue in both the energy market and the ramp-up market. To address the uncertainty of the bidding probability in the ramp-up market, information gap decision theory is introduced, transforming the model into an IGDT robust model that aims to maximize uncertainty tolerance while ensuring minimum profit. This allows the final decision to guarantee revenue security even in the most unfavorable market conditions, achieving controllable risk. By transforming the complex IGDT robust model into an efficiently solvable mixed-integer linear programming problem, bidding decisions are quickly obtained. Finally, the bidding decision is decomposed into specific charging and discharging power commands for each scenario and executed, forming a safe closed-loop control from optimization decision to physical execution, ensuring its feasibility and safety when participating in grid regulation. Attached Figure Description

[0027] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 A flowchart illustrating a collaborative scheduling method for electric vehicle clusters oriented towards multi-scenario vehicle-network interaction provided in an embodiment of this application; Figure 2 A schematic diagram of the overall framework of the electric vehicle cluster collaborative scheduling method for multi-scenario vehicle-network interaction provided in the embodiments of this application; Figure 3 A schematic diagram of basic battery parameters for various scenarios provided in the embodiments of this application; Figure 4 A schematic diagram illustrating the maximum adjustable capacity of a multi-scenario electric vehicle provided in this application embodiment; Figure 5 A schematic diagram illustrating the maximum charging and discharging power of electric vehicles in various scenarios provided in this application embodiment; Figure 6 A schematic diagram illustrating the benchmark electricity prices for the day-ahead market, real-time market, and ramp-up market provided for embodiments of this application; Figure 7 A schematic diagram illustrating the daily revenue for each scenario under different cases provided in the embodiments of this application; Figure 8 A schematic diagram of the day-ahead market charging and discharging power for residential scenarios provided in this application embodiment; Figure 9 A schematic diagram of the real-time market charging and discharging power in a residential community scenario provided in this application embodiment; Figure 10 The application capacity for the hill-climbing market in the bus fleet scenario provided in this application embodiment. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0029] To address the challenges of regulating distributed electric vehicle (EV) clusters with high uncertainty and complex physical constraints, ensuring their feasibility and safety in participating in grid regulation, this application provides a collaborative scheduling method for EV clusters in multiple vehicle-grid interaction scenarios. Please refer to... ξ This is a flowchart illustrating a method for collaborative scheduling of electric vehicle clusters for multi-scenario vehicle-to-grid interaction provided in this application embodiment. The following is a description of... ω The method for collaborative scheduling of electric vehicle clusters for multi-scenario vehicle-network interaction is introduced.

[0030] S101. Adjustable capability modeling is performed for various typical electric vehicle charging scenarios, and the maximum charging and discharging power and maximum adjustable energy of each scenario are calculated.

[0031] Multiple scenarios are included, such as residential community charging, centralized commercial charging, bus fleet charging, urban logistics, and heavy-duty truck battery swapping. The modeling process for each scenario is described below.

[0032] The first type is the residential community application scenario.

[0033] S1.1. Use beta distribution to simulate the arrival and departure times of each resident vehicle, and define the on-site status of each resident vehicle at each time.

[0034]

[0035] in, For the first i The access time of a resident vehicle; Beta ( α R , β R ) indicates that the shape parameter is α R and β RThe beta distribution; access times are distributed between 17:00 and 23:00 in the evening; For the first i The departure time of the resident vehicle; Beta ( ζ R , η R ) indicates that the shape parameter is where Q R and η R The beta distribution; the departure time is between 06:00 and 14:00 in the morning.

[0036]

[0037] in, For the first i Residential vehicles at time t The status of being on the station.

[0038] S1.2. The initial state of charge (SOC) of each resident vehicle is simulated using a truncated normal distribution.

[0039]

[0040] η R,i Let i be the initial state of charge of the i-th resident vehicle; The mean is μ R variance is The truncated normal distribution.

[0041] S1.3. Based on the on-site status, the initial state of charge of each resident vehicle, battery capacity parameters, and charging facility capacity parameters, calculate the maximum adjustable energy and the maximum charging and discharging power.

[0042]

[0043] In the formula, Q R,t The maximum adjustable energy for residential use within the community; P R,t The maximum charging and discharging power for residential use in the community; N R The number of vehicles in the community; γ R,i Let i be the initial state of charge of the i-th resident vehicle; This is the lower limit for the battery capacity of residential vehicles; C R,i Let be the battery capacity of the i-th resident vehicle; Indicates the time of the i-th resident vehicle t The on-site status; The maximum charging and discharging power of the i-th resident vehicle; The number of available charging stations in the community; represents the maximum charging and discharging power of the cell; min represents the minimum value function.

[0044] In this embodiment, a beta distribution is used to simulate access / departure times, and a truncated normal distribution is used to simulate the initial state of charge (SOC). The beta distribution can effectively simulate the concentrated behavior of residents returning home at night and charging at night, making the predicted vehicle on-site status more realistic and avoiding the biases caused by uniform distribution or simple time windows. Combined with the distribution of the initial SOC, the maximum charging and discharging power and maximum adjustable energy of residential areas during nighttime peak load periods can be calculated more accurately, providing precise data support for the implementation of orderly charging in residential areas.

[0045] The second type is the centralized commercial charging scenario.

[0046] S2.1. The number of arriving vehicles at each time step is simulated using a Poisson distribution.

[0047]

[0048] For a moment t The number of vehicles arriving at the centralized charging station; Poisson(λ) M ) indicates that the strength parameter is λ M The Poisson distribution.

[0049] S2.2. The initial state of charge of each commercial charging vehicle is simulated using a triangular distribution.

[0050]

[0051] Figure 2 (M,j) For the first j The initial state of charge of a commercial vehicle; Triangular ( a M ,b M ,c M ) indicates that the lower limit is a M The upper limit is b M The mode is c M The triangular distribution.

[0052] S2.3 Define the state variables for each commercial charging vehicle connected to the charging pile.

[0053]

[0054] ΨM,j,t For the first j The status variable for a commercial vehicle connected to a charging pile: 1 indicates connection, and 0 indicates no connection.

[0055] S2.4. Based on the number of vehicles arriving at the station, the initial state of charge of each commercial charging vehicle, state variables, battery capacity parameters, and charging facility capacity parameters, calculate the maximum adjustable energy and the maximum charging and discharging power.

[0056]

[0057] In the formula, Q M,t To maximize the adjustable energy for centralized commercial charging scenarios; P M,t For centralized commercial charging scenarios, the maximum charging and discharging power is required; M,t For a moment t The number of vehicles arriving at centralized charging stations; Figure 3 M,j For the first j The initial state of charge of the commercial charging vehicle; The lower limit of charging power for commercial vehicles; C M,j The battery capacity of the j-th commercial vehicle; Ψ M,j,t For the first j State variables of commercial vehicles connected to charging piles; For the first j The maximum power of the commercial charging vehicle; The number of available charging piles at the centralized charging station; represents the maximum charging and discharging power within the centralized charging station; min represents the minimum value function.

[0058] In this embodiment, the Poisson distribution is used to simulate the number of vehicles arriving per unit time, and the triangular distribution is used to simulate the initial SOC. The Poisson distribution can effectively describe the random and independent arrival characteristics of commercial charging vehicles, and can accurately predict the concentrated arrival peaks that may occur during lunch breaks and evenings. The triangular distribution (based on the minimum, most likely, and maximum values) is suitable for quickly and reasonably estimating variables such as commercial charging user SOC, where information is incomplete but the range is known.

[0059] The third scenario is a bus fleet.

[0060] S3.1. The number of buses in the station at each time point is simulated using a Poisson distribution.

[0061]

[0062] ν B,t Let be the number of buses at station t; Poisson(λ) B ) is the strength parameter λ BThe Poisson distribution.

[0063] S3.2 Define the station status variable for each bus based on a fixed entry and exit timetable.

[0064]

[0065] Y B,k,t Let be the station-state variable of the k-th bus at time t; Let k be the arrival time of the kth bus. Let k be the departure time of the kth bus.

[0066] S3.3. The initial state of charge of each bus is simulated using a truncated normal distribution.

[0067]

[0068] ζ B,k Let K be the initial state of charge of the k-th bus. The mean is μ B variance is The truncated normal distribution.

[0069] S3.4. Based on the number of buses, on-site state variables, initial state of charge of each bus, battery capacity parameters, and charging facility capacity parameters, calculate the maximum adjustable energy and maximum charging and discharging power.

[0070]

[0071] In the formula, Q B,t The maximum adjustable energy for bus fleet scenarios; P B,t Maximum charging and discharging power for bus fleet scenarios; ν B,t The number of buses at station t; Figure 4 B,k Let K be the initial state of charge of the k-th bus. This is the minimum battery level for buses; q B,k Y represents the battery capacity of the kth bus; B,k,t Let be the station-state variable of the k-th bus at time t; The maximum charging and discharging power of the k-th bus; The number of charging parking spaces at bus stops; The maximum charging and discharging power of the bus station is denoted by ; min represents the minimum value function.

[0072] In this embodiment, the station status is determined based on a fixed timetable, the number of vehicles in the station is simulated using a Poisson distribution, and the initial State of Charge (SOC) is simulated using a truncated normal distribution. The "fixed time window" ensures the basic adjustable time period based on the timetable, while the "Poisson distribution" quantifies the fluctuations in the actual number of vehicles in the station due to road congestion and other factors. This hybrid model combines the reliability of planning with the realism of randomness, thereby accurately predicting the maximum adjustable energy and maximum charging / discharging power of the bus fleet. The fourth scenario is urban distribution logistics.

[0073] S4.1. The Poisson distribution is used to simulate the number of delivery batches arriving per unit time, and the log-normal distribution is used to simulate the number of logistics vehicles arriving in each batch.

[0074]

[0075] In the formula, For a moment t Number of arrival batches; Poisson(λ) L ) represents an intensity of λ L The Poisson distribution; Figure 5 L,t For a moment t The number of logistics vehicles arriving in each batch; Lognormal(μ L ,σ L ) indicates that the mean is μ L The variance is σ L It follows a log-normal distribution.

[0076] S4.2. The initial state of charge of each logistics vehicle is simulated using a uniform distribution.

[0077]

[0078] In the formula, For the first m The initial state of charge of the logistics vehicle; Indicates interval [ A uniform distribution on [the surface].

[0079] S4.1 Calculate the maximum adjustable energy and maximum charge / discharge power based on the number of logistics vehicles, the initial state of charge of each logistics vehicle, the battery capacity parameters, and the charging facility capacity parameters.

[0080]

[0081] In the formula, Q L,t The maximum adjustable energy for urban distribution logistics scenarios; P L,t This represents the maximum charging and discharging power for urban logistics scenarios. For the firstm The initial state of charge of the logistics vehicle; Figure 6 L,t For a moment t The number of logistics vehicles arriving at the station in each batch; This is the minimum battery capacity for logistics vehicles; Φ L,m,t For the first m The state variables of a logistics vehicle; C L,m For the first m Battery capacity of a logistics vehicle; For the first m The maximum charging and discharging power of the logistics vehicle; The number of charging stations at the logistics station; The maximum charging and discharging power of the logistics station is denoted by ; min represents the minimum value function.

[0082] In this embodiment, a composite probability distribution (Poisson distribution simulating the number of arriving batches + log-normal distribution simulating the number of vehicles per batch) is used to uniformly simulate the initial State of Charge (SOC), accurately depicting the core characteristic of logistics parks: "batch arrivals of varying sizes." The log-normal distribution effectively simulates the random fluctuations in the number of vehicles per batch (which may be small or large), thus more realistically reflecting the sharp rises and falls in the load of the logistics center. The behavior pattern of logistics vehicles charging during delivery breaks is clarified, effectively leveraging their adjustable capabilities across multiple fixed time periods during the day.

[0083] The fifth scenario is battery swapping for heavy-duty trucks.

[0084] S5.1 Calculate the expected number of available batteries based on the queuing theory model.

[0085]

[0086] In the formula, For a moment t Expected number of available batteries; S H This is the upper limit for battery inventory at heavy-duty truck battery swapping stations; π H,u,t For a moment t There are on the station u The probability of a block battery. For heavy-duty truck battery swapping station arrival rate; For service rate; Figure 7 H This is the ratio of the arrival rate to the service rate of heavy-duty truck battery swapping stations.

[0087] S5.2. The initial state of charge of each backup battery is simulated using beta distribution.

[0088]

[0089] This represents the initial state of charge of the nth battery. Indicates shape parameters as a H and b H The beta distribution.

[0090] S5.3 Calculate the adjustable energy and maximum charge / discharge power based on the expected number of available batteries, the initial state of charge of each backup battery, battery capacity parameters, and charging facility capacity parameters.

[0091]

[0092] Figure 8 H,t The maximum adjustable energy for heavy-duty truck battery swapping scenarios; P H,t This represents the maximum charging and discharging power for heavy-duty truck battery swapping scenarios. For a moment t The expected number of available batteries; This represents the initial state of charge of the nth battery. This is the lower limit of the battery capacity for heavy-duty truck battery swapping stations; C H,n Let n be the capacity of the nth battery. This refers to the number of battery swapping locations; This is the maximum power of the heavy-duty truck battery swapping station; denoted as the maximum charge / discharge power of the nth battery; min represents the minimum value function.

[0093] In this embodiment, a queuing theory model is introduced, treating the heavy-duty truck battery swapping station as a dynamic service system. Its adjustability is assessed by calculating the expected number of available batteries. This accurately reflects the flexible nature of the heavy-duty truck battery swapping station as a "shared battery," solving the assessment challenges arising from its service process and battery turnover characteristics, and making the assessment results more consistent with actual operating conditions.

[0094] S102. Based on the maximum charging and discharging power and maximum adjustable energy of each scenario, construct constraints and, with the goal of maximizing the comprehensive revenue of electric vehicle aggregators in the electric energy market and the ancillary services market, construct a two-stage optimization model that includes the day-ahead market and the real-time market.

[0095] In one possible embodiment, the constraints include power constraints, regulation capacity constraints, and energy dynamic constraints.

[0096] The power constraints are as follows:

[0097] In the formula, P s,t,max The maximum charge / discharge power at time t in scenario s; This represents the day-ahead market charging level at time t in scenario s; This represents the day-ahead market discharge volume at time t in scenario s; This represents the application volume of the climbing market per unit time at time t in scenario s; express s In the scene t The number of applications submitted in the downhill market per unit of time.

[0098] The capacity adjustment constraints are as follows:

[0099] In the formula, This represents the application volume of the climbing market per unit time at time t in scenario s; express s In the scene t The application volume of the downhill market per unit time period; The minimum application capacity for the climbing market; This is the minimum application capacity for the downhill market.

[0100] The energy dynamic constraints are as follows:

[0101] In the formula, This represents the application volume of the climbing market per unit time at time t in scenario s; express s In the scene t The application volume of the downhill market per unit time period; Figure 9 cha Indicates charging efficiency; Figure 10 dis Indicates discharge efficiency; Δt represents unit time; SOC min E represents the lowest initial state of charge; s express t Battery energy at any moment; E s,t,max express s In the scene t The maximum adjustable energy at any given time.

[0102] In one possible embodiment, the function of the overall benefit is as follows:

[0103] Where max represents the maximization function; Q represents the overall return; This indicates the revenue from electricity generated by participating in the day-ahead market; This indicates the revenue from participating in the real-time market for electrical energy; This indicates the returns from participating in the uphill market; Indicates the returns from participating in the downhill market; C deg This represents the cost of battery degradation caused by charging and discharging operations in electric vehicles.

[0104] The formulas for calculating various benefits and battery degradation costs are as follows:

[0105] in, express t The electricity price in the market at that moment; express t Real-time market electricity prices at any given moment; This represents the day-ahead market charging level at time t in scenario s; This represents the day-ahead market discharge volume at time t in scenario s; express s In the scene t Real-time market charging power; express s In the scene t Real-time market discharge volume at any given moment; This indicates the probability of winning a bid in the uphill market; This indicates the probability of winning a bid in the downhill market. express t Prices in the market are constantly climbing; express t The current price in the ramp-up market; express s In the scene t The application volume of the climbing market per unit time period; express s In the scene t The application volume of the downhill market per unit time period; The degradation cost per unit of electricity.

[0106] In this embodiment, instead of unilaterally pursuing short-term market gains while ignoring hardware losses, the battery degradation cost is incorporated as a core deduction item into the objective function. This design allows the optimization algorithm to automatically weigh the "market gains from a single charge-discharge operation" against the "resulting in battery life loss" when making decisions. This effectively avoids behaviors that damage battery health, such as frequent and deep charge-discharge operations for meager profits, and guides a smoother and more scientific charge-discharge plan. Thus, while pursuing maximum profits, it ensures the safety and lifespan of the electric vehicle cluster as a core asset, achieving a balance between long-term and short-term economic efficiency.

[0107] S103. Information gap decision theory (IGDT) is used to quantify the uncertainty of the winning probability of the market capacity ramp-up, and the two-stage stochastic optimization model is transformed into an IGDT robust model.

[0108] In one possible embodiment, the specific steps of S103 are as follows: S6.1 Define the radius of uncertainty.

[0109] The radius of uncertainty includes the first radius of uncertainty parameter. α 1 and second uncertain radius parameters α 2; The value of the uncertainty radius is between [0,1]. First uncertainty radius parameter α 1 represents the relative deviation of the probability of winning a bid in the uphill market; the second uncertainty radius parameter. α 2 represents the relative deviation of the probability of winning a bid in the downhill market.

[0110] S6.2. Set the first and second uncertain radius parameters to zero. When the winning probability in the climbing market is the benchmark value, solve the two-stage stochastic optimization model to obtain the benchmark profit.

[0111]

[0112] z represents the benchmark profit; max represents the maximization function. Q Indicates overall return; This indicates the probability of winning a bid in the uphill market; This indicates the probability of winning a bid in the downhill market. x This represents the set of decision variables in the day-to-day market, including the day-to-day charging and discharging volume (i.e., the amount of electricity charged and discharged in the day-to-day market). and ) and the declared capacity of the uphill and downhill markets (i.e. and ); y This represents the set of decision variables in the real-time market, including the real-time charging and discharging capacity (i.e., the amount of electricity charged and discharged in the real-time market). and ).

[0113] make α 1 and α 2 equals 0, thus obtaining the benchmark profit. z .

[0114] S6.3. Based on the benchmark profit, set an acceptable minimum profit.

[0115] S6.4. With the goal of maximizing the tolerance for uncertainty and the constraint that the overall return is not less than the minimum profit, construct an IGDT robust model.

[0116] The robust IGDT model is as follows:

[0117] Where max represents the maximization function; α 1 represents the first uncertain radius parameter; α 2 represents the second uncertain radius parameter; This indicates the probability of winning a bid in the uphill market; This indicates the probability of winning a bid in the downhill market. ξ Indicates the profit loss rate; z represents the benchmark profit; Q represents the overall return. x This represents the set of decision variables in the current market. y This represents the set of decision variables in a real-time market.

[0118] In this embodiment, the IGDT method simulates the range of uncertainty by continuously adjusting the radius parameter, and ensures that the profit does not fall below a certain set lower limit even in the worst case, thereby obtaining a robust optimal solution. Traditional stochastic optimization methods heavily rely on the precise probability distribution of uncertain parameters. However, in emerging climbing markets, historical data is scarce, making it difficult to accurately obtain their probability distributions. The IGDT method does not rely on the probability distribution information of uncertain quantities; it only needs to know their baseline value and possible deviation directions. This perfectly solves the decision-making problem in situations with insufficient data or changing market rules.

[0119] S104. Transform the IGDT robust model into a mixed-integer linear programming problem and solve it to obtain the bidding decision that can still guarantee the minimum profit under the most unfavorable uncertainty.

[0120] Specifically, the IGDT robust model is transformed into a mixed integer linear programming (MLP) problem model, and a mathematical programming solver is used to solve the MILP model. The output bidding decision includes the following: 1. The set of decision variables x in the current day's market, including the current day's charging volume. The current market discharge volume Declaration capacity of the uphill market Declaration capacity of the downhill market .

[0121] 2. The set of decision variables y in the real-time market, including the real-time charging volume. Real-time market discharge volume .

[0122] 3. Uncertain radius α 1 and α 2.

[0123] S105. Based on the bidding decision, generate charging and discharging power plan instructions for each scenario and issue them to the charging facilities in each scenario for execution.

[0124] Specifically, based on the set of decision variables in bidding decisions x and y The system generates a global, overall charging and discharging power plan curve, decomposes the total power plan into various electric vehicle charging scenarios, and generates standardized charging and discharging power plan instructions for each scenario. These instructions are typically a sequence containing timestamps and power values. Through a secure and reliable communication network, the formatted instructions are sent to the central controller of the charging facility or the charging pile group management system for the corresponding scenario. Upon receiving the instructions, the local controllers in each scenario further allocate the power instructions to specific charging piles or battery swapping devices according to their internal logic. The charging piles execute the instructions, precisely controlling their output power to achieve charging (drawing power from the grid), discharging (feeding power to the grid, V2G), or standby mode.

[0125] Please refer to ω This diagram illustrates the overall framework of the electric vehicle cluster collaborative scheduling method for multi-scenario vehicle-to-grid interaction proposed in this application. It mainly consists of two core models: one is the adjustable capacity modeling for multiple typical scenarios, and the other is the two-stage bidding decision optimization for multiple markets. The multi-scenario adjustable capacity modeling analyzes battery characteristics, driver behavior, and charging capacity constraints in five typical charging scenarios: residential charging, centralized commercial charging, bus fleets, urban delivery, and heavy-duty truck battery swapping, providing boundary conditions for the market bidding optimization model. The two-stage bidding decision optimization for multiple markets uses maximizing the comprehensive revenue of EV aggregators across multiple markets as the objective function. It optimizes the EV resource bidding strategy in the day-ahead phase and smooths out deviations in the real-time phase. It also introduces IGDT to quantify ramp-up market prices and bidding uncertainty, constructing a mixed-integer linear programming model to obtain the EV bidding decisions in the energy-ramp-up market. Finally, based on these bidding decisions, charging and discharging commands for each scenario are generated and issued.

[0126] To verify the effectiveness of the method proposed in this application, a simulation case analysis was conducted using electric vehicle operation data from a certain region and historical price data from the US PJM electricity market. The case parameter settings are as follows: ζ As shown, the maximum adjustable capacity (i.e., maximum adjustable energy) of electric vehicles in multiple scenarios is as follows: η As shown, the maximum adjustable power of electric vehicles in multiple scenarios is as follows: where Q As shown, the benchmark electricity prices for the day-ahead market, real-time market, and ramp-up market are as follows: η As shown.

[0127] To compare the profitability of different market participation models, three case studies were set up for comparative analysis, including: Case 1: Only participate in the day-ahead market, formulate charging and discharging strategies based on the day-ahead price, and realize peak-valley arbitrage.

[0128] Case 2: Simultaneously participate in both the day-ahead and real-time markets, and optimize for deviations based on the day-ahead decision and the price fluctuations in the real-time market to achieve two-level price arbitrage.

[0129] Case 3: Building on Case 2, this case further introduces the ramp-up assistance service market, which participates in the application for ramp-up and ramp-down capacity, and uses the IGDT robust model proposed in this application for overall decision-making.

[0130] Daily revenue for different scenarios in different cases, such as η As shown, the day-to-day charging and discharging power of residential applications in residential communities is as follows: γ As shown, the real-time charging and discharging power of the residential market in the community scenario is as follows: ​ As shown, the application capacity of the ramp-up market in the bus fleet scenario is as follows: ​ As shown.

[0131] In summary, this application provides a collaborative scheduling method for electric vehicle (EV) clusters in multiple scenarios of vehicle-to-grid interaction. It constructs adjustable EV capabilities across various scenarios, accurately calculating the differentiated adjustment characteristics of EV clusters under different charging scenarios, including five typical scenarios: residential areas, commercial charging stations, bus depots, urban logistics, and heavy-duty truck battery swapping stations. This provides scheduling potential boundaries for market bidding. Simultaneously, by optimizing EV participation in the ramp-up market bidding decision through IGDT, it can handle the uncertainties of electricity prices and winning probabilities at both the day-ahead and real-time stages, maximizing risk tolerance while ensuring baseline returns. Finally, based on the bidding decisions, it generates charging and discharging power plan instructions for each scenario.

[0132] Based on the same inventive concept, this application also provides an electric vehicle cluster collaborative scheduling device for multi-scenario vehicle-to-grid interaction, the device comprising: The multi-scenario modeling module is used to model the adjustable capabilities of various typical electric vehicle charging scenarios and calculate the maximum charging and discharging power and maximum adjustable energy for each scenario. These scenarios include residential charging scenarios, centralized commercial charging scenarios, bus fleet scenarios, urban logistics scenarios, and heavy truck battery swapping scenarios. The model building module is used to construct constraints based on the maximum charging and discharging power and maximum adjustable energy in each scenario, with the goal of maximizing the comprehensive revenue of electric vehicle aggregators in the electric energy market and the ancillary services market, and to build a two-stage optimization model that includes the day-ahead market and the real-time market. The model optimization module is used to quantify the uncertainty of the winning probability of the market capacity ramp-up using the Information Gap Decision Theory (IGDT), and transform the two-stage stochastic optimization model into an IGDT robust model. The decision module is used to transform the IGDT robust model into a mixed integer linear programming problem for solution, and obtain the bidding decision that can still guarantee the minimum profit under the most unfavorable uncertainty. The collaborative scheduling module is used to generate charging and discharging power plan instructions for each scenario based on the bidding decision, and then issue them to the charging facilities in each scenario for execution.

[0133] Optionally, the multi-scene modeling module is specifically used for: For residential scenarios, beta distribution is used to simulate the access and departure times of each resident vehicle, and the on-site status of each resident vehicle at each time is defined. The initial state of charge of each resident vehicle was simulated using a truncated normal distribution; Based on the on-site status, the initial state of charge of each resident vehicle, battery capacity parameters, and charging facility capacity parameters, the maximum adjustable energy and maximum charge / discharge power are calculated using the following formulas:

[0134] In the formula, Q R,t The maximum adjustable energy for residential use within the community; P R,t The maximum charging and discharging power for residential use in the community; N R The number of vehicles in the community; ​ R,i Let i be the initial state of charge of the i-th resident vehicle; C is the lower limit for the battery capacity of residential vehicles; R,i Let be the battery capacity of the i-th resident vehicle; Indicates the time of the i-th resident vehicle t The on-site status; The maximum charging and discharging power of the i-th resident vehicle; The number of available charging stations in the community; represents the maximum charging and discharging power of the cell; min represents the minimum value function.

[0135] Optionally, the multi-scene modeling module is specifically used for: For centralized commercial charging scenarios, a Poisson distribution is used to simulate the number of arriving vehicles at each time step; The initial state of charge of each commercial charging vehicle is simulated using a triangular distribution. Define the state variables for each commercial charging vehicle connected to the charging pile; Based on the number of arriving vehicles, the initial state of charge of each commercial charging vehicle, state variables, battery capacity parameters, and charging facility capacity parameters, the maximum adjustable energy and maximum charging / discharging power are calculated using the following formulas:

[0136] In the formula, Q M,t To maximize the adjustable energy for centralized commercial charging scenarios; PM,t For centralized commercial charging scenarios, the maximum charging and discharging power is required; M,t For a moment t The number of vehicles arriving at centralized charging stations; ​ M,j For the first j The initial state of charge of the commercial charging vehicle; The lower limit of charging power for commercial vehicles; C M,j The battery capacity of the j-th commercial vehicle; Ψ M,j,t For the first j State variables of commercial vehicles connected to charging piles; For the first j The maximum power of the commercial charging vehicle; The number of available charging piles at the centralized charging station; represents the maximum charging and discharging power within the centralized charging station; min represents the minimum value function.

[0137] Optionally, the multi-scene modeling module is specifically used for: For the bus fleet scenario, a Poisson distribution is used to simulate the number of buses in the station at each time step; The station status variable for each bus is defined based on a fixed entry and exit timetable; The initial state of charge of each bus was simulated using a truncated normal distribution; Based on the number of buses, on-site state variables, initial state of charge of each bus, battery capacity parameters, and charging facility capacity parameters, the maximum adjustable energy and maximum charge / discharge power are calculated using the following formulas:

[0138] In the formula, Q B,t The maximum adjustable energy for bus fleet scenarios; P B,t Maximum charging and discharging power for bus fleet scenarios; ν B,t The number of buses at station t; ​ B,k Let K be the initial state of charge of the k-th bus. This is the minimum battery level for buses; q B,k Y represents the battery capacity of the kth bus; B,k,t Let be the station-state variable of the k-th bus at time t; The maximum charging and discharging power of the k-th bus; The number of charging parking spaces at bus stops; The maximum charging and discharging power of the bus station is denoted by ; min represents the minimum value function.

[0139] Optionally, the multi-scene modeling module is specifically used for: For urban distribution logistics scenarios, Poisson distribution is used to simulate the number of delivery batches arriving per unit time, and log-normal distribution is used to simulate the number of logistics vehicles arriving in each batch. The initial state of charge of each logistics vehicle is simulated using a uniform distribution. Based on the number of logistics vehicles, the initial state of charge of each vehicle, battery capacity parameters, and charging facility capacity parameters, the maximum adjustable energy and maximum charge / discharge power are calculated using the following formulas:

[0140] In the formula, Q L,t The maximum adjustable energy for urban distribution logistics scenarios; P L,t This represents the maximum charging and discharging power for urban logistics scenarios. For the first m The initial state of charge of the logistics vehicle; ​ L,t For a moment t The number of logistics vehicles arriving at the station in each batch; This is the minimum battery capacity for logistics vehicles; Φ L,m,t For the first m The state variables of a logistics vehicle; C L,m For the first m Battery capacity of a logistics vehicle; For the first m The maximum charging and discharging power of the logistics vehicle; The number of charging stations at the logistics station; The maximum charging and discharging power of the logistics station is denoted by ; min represents the minimum value function.

[0141] Optionally, the multi-scene modeling module is specifically used for: For heavy-duty truck battery swapping scenarios, the expected number of available batteries is calculated based on a queuing theory model; The initial state of charge of each backup battery is simulated using a beta distribution. Based on the expected number of available batteries, the initial state of charge of each backup battery, battery capacity parameters, and charging facility capacity parameters, the adjustable energy and maximum charge / discharge power are calculated using the following formulas:

[0142] ​ H,t The maximum adjustable energy for heavy-duty truck battery swapping scenarios; P H,t This represents the maximum charging and discharging power for heavy-duty truck battery swapping scenarios. For a moment t The expected number of available batteries; This represents the initial state of charge of the nth battery. C is the lower limit of the battery capacity for heavy-duty truck battery swapping stations; H,n Let n be the capacity of the nth battery. This refers to the number of battery swapping locations; This is the maximum power of the heavy-duty truck battery swapping station; denoted as the maximum charge / discharge power of the nth battery; min represents the minimum value function.

[0143] Optional, the function for calculating the overall return is as follows:

[0144] Where max represents the maximization function; Q represents the overall return; This indicates the revenue from electricity generated by participating in the day-ahead market; This indicates the revenue from participating in the real-time market for electrical energy; This indicates the returns from participating in the uphill market; Indicates the returns from participating in the downhill market; C deg This represents the cost of battery degradation caused by charging and discharging operations in electric vehicles.

[0145] Optional constraints include power constraints, regulation capacity constraints, and energy dynamic constraints; The power constraints are as follows:

[0146] The capacity adjustment constraints are as follows:

[0147] The energy dynamic constraints are as follows:

[0148] In the formula, This represents the day-ahead market charging level at time t in scenario s; This represents the day-ahead market discharge volume at time t in scenario s; This represents the application volume of the climbing market per unit time at time t in scenario s; express s In the scene t The application volume of the downhill market per unit time; P s,t,max For scene s t Maximum charging and discharging power at any given moment; express s In the scene t Real-time market charging power; express s In the scene t Real-time market discharge volume at any given moment; This indicates the minimum application capacity for the climbing market; This indicates the minimum application capacity for the downhill market; ​ cha Indicates charging efficiency; ​ dis Indicates discharge efficiency; Δt represents unit time; SOC min E represents the lowest initial state of charge; s express t Battery energy at any moment; E s,t,max express s In the scene t The maximum adjustable energy at any given time.

[0149] Optionally, the model optimization module is specifically used for: Define the radius of uncertainty; the radius of uncertainty includes a first radius of uncertainty parameter and a second radius of uncertainty parameter; the first radius of uncertainty parameter represents the relative deviation of the probability of winning the bid in the uphill market; the second radius of uncertainty parameter represents the relative deviation of the probability of winning the bid in the downhill market. Set the first and second uncertainty radius parameters to zero, and when the winning probability in the climbing market is the benchmark value, solve the two-stage stochastic optimization model to obtain the benchmark profit. Based on the benchmark profit, set an acceptable minimum profit; With the goal of maximizing the tolerance for uncertainty and the constraint that the overall return is no less than the minimum profit, an IGDT robust model is constructed.

[0150] Optional, the robust model for IGDT is as follows:

[0151] Where max represents the maximization function; α 1 represents the first uncertain radius parameter; α 2 represents the second uncertain radius parameter; This indicates the probability of winning a bid in the uphill market; This indicates the probability of winning a bid in the downhill market. ​ Indicates the profit loss rate; z Indicates benchmark profit; Q Indicates overall return; x This represents the set of decision variables in the current market. y This represents the set of decision variables in a real-time market.

[0152] It should be noted that each module in the electric vehicle cluster collaborative scheduling device for multi-scenario vehicle-network interaction in this embodiment corresponds one-to-one with each step in the electric vehicle cluster collaborative scheduling method for multi-scenario vehicle-network interaction in the aforementioned embodiment. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned electric vehicle cluster collaborative scheduling method for multi-scenario vehicle-network interaction, and will not be repeated here.

[0153] Based on the same inventive concept, this application also provides a computer device, which includes a processor, a memory, and a computer program stored in the memory. The computer program is executed by the processor to implement the aforementioned electric vehicle cluster collaborative scheduling method for multi-scenario vehicle-to-grid interaction.

[0154] Based on the same inventive concept, this application also provides a computer storage medium storing a computer program, which is executed by a processor to implement the aforementioned electric vehicle cluster collaborative scheduling method for multi-scenario vehicle-to-grid interaction.

[0155] In some embodiments, the computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; or it may be a device including one or any combination of the above-mentioned memories. The computer may be a variety of computing devices, including smart terminals and servers.

[0156] In some embodiments, executable instructions may take the form of a program, software, software module, script, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a standalone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0157] As an example, executable instructions may, but do not necessarily, correspond to files in the file system. They may be stored as part of a file that holds other programs or data, for example, in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple co-located files (e.g., a file that stores one or more modules, subroutines, or code sections).

[0158] As an example, executable instructions can be deployed to execute on a single computing device, or on multiple computing devices located in one location, or on multiple computing devices distributed across multiple locations and interconnected via a communication network.

[0159] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0160] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0161] The above specific embodiments further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for collaborative scheduling of electric vehicle clusters for multi-scenario vehicle-to-grid interaction, characterized in that, include: Adjustable capability models were created for various typical electric vehicle charging scenarios, and the maximum charging and discharging power and maximum adjustable energy for each scenario were calculated. These scenarios include residential charging scenarios, centralized commercial charging scenarios, bus fleet scenarios, urban logistics scenarios, and heavy truck battery swapping scenarios. Based on the maximum charging and discharging power and maximum adjustable energy in each scenario, constraints are constructed to maximize the comprehensive revenue of electric vehicle aggregators in the electric energy market and the ancillary services market. A two-stage optimization model including the day-ahead market and the real-time market is then constructed. The uncertainty of the winning probability of the market capacity ramp-up is quantified by using the information gap decision theory (IGDT), and the two-stage stochastic optimization model is transformed into an IGDT robust model. The robust IGDT model is transformed into a mixed-integer linear programming problem for solution, resulting in a bidding decision that guarantees the minimum profit even under the most unfavorable uncertainty. Based on the bidding decision, charging and discharging power plan instructions for each scenario are generated and issued to the charging facilities in each scenario for execution.

2. The electric vehicle cluster collaborative scheduling method for multi-scenario vehicle-to-grid interaction according to claim 1, characterized in that, The adjustable capability modeling for various typical electric vehicle charging scenarios is performed, and the maximum charging / discharging power and maximum adjustable energy for each scenario are calculated, including: For the residential community scenario, beta distribution is used to simulate the access time and departure time of each resident vehicle, and the on-site status of each resident vehicle at each time is defined; The initial state of charge of each resident vehicle was simulated using a truncated normal distribution; Based on the on-site status, the initial state of charge of each resident vehicle, battery capacity parameters, and charging facility capacity parameters, the maximum adjustable energy and maximum charge / discharge power are calculated using the following formulas: ； In the formula, Q R,t This represents the maximum adjustable energy for the residential application scenario in the community. P R,t This refers to the maximum charging and discharging power for residential use in the aforementioned community. N R The number of vehicles in the community; ξ R,i Let i be the initial state of charge of the i-th resident vehicle; C is the lower limit for the battery capacity of residential vehicles; R,i Let be the battery capacity of the i-th resident vehicle; Indicates the time of the i-th resident vehicle t The on-site status; The maximum charging and discharging power of the i-th resident vehicle; The number of available charging stations in the community; represents the maximum charging and discharging power of the cell; min represents the minimum value function.

3. The electric vehicle cluster collaborative scheduling method for multi-scenario vehicle-to-grid interaction according to claim 1, characterized in that, The adjustable capability modeling for various typical electric vehicle charging scenarios is performed, and the maximum charging / discharging power and maximum adjustable energy for each scenario are calculated, including: For the centralized commercial charging scenario, a Poisson distribution is used to simulate the number of arriving vehicles at each time step; The initial state of charge of each commercial charging vehicle is simulated using a triangular distribution. Define the state variables for each commercial charging vehicle connected to the charging pile; Based on the number of arriving vehicles, the initial state of charge of each commercial charging vehicle, the state variables, battery capacity parameters, and charging facility capacity parameters, the maximum adjustable energy and maximum charging / discharging power are calculated using the following formulas: ； In the formula, Q M,t This represents the maximum adjustable energy for the centralized commercial charging scenario. P M,t The maximum charging and discharging power for the aforementioned centralized commercial charging scenario; κ M,t For a moment t The number of vehicles arriving at centralized charging stations; ω M,j For the first j The initial state of charge of the commercial charging vehicle; The lower limit of charging power for commercial vehicles; C M,j The battery capacity of the j-th commercial vehicle; Ψ M,j,t For the first j State variables of commercial vehicles connected to charging piles; For the first j The maximum power of the commercial charging vehicle; The number of available charging piles at the centralized charging station; represents the maximum charging and discharging power within the centralized charging station; min represents the minimum value function.

4. The electric vehicle cluster collaborative scheduling method for multi-scenario vehicle-to-grid interaction according to claim 1, characterized in that, The adjustable capability modeling for various typical electric vehicle charging scenarios is performed, and the maximum charging / discharging power and maximum adjustable energy for each scenario are calculated, including: For the bus fleet scenario, a Poisson distribution is used to simulate the number of buses at each time point. The station status variable for each bus is defined based on a fixed entry and exit timetable; The initial state of charge of each bus was simulated using a truncated normal distribution; Based on the number of buses, the on-site state variables, the initial state of charge of each bus, battery capacity parameters, and charging facility capacity parameters, the maximum adjustable energy and maximum charging / discharging power are calculated using the following formulas: ； In the formula, Q B,t This represents the maximum adjustable energy for the bus fleet scenario. P B,t The maximum charging and discharging power for the bus fleet scenario; ν B,t The number of buses at station t; ζ B,k Let K be the initial state of charge of the k-th bus. This is the minimum battery level for buses; q B,k Y represents the battery capacity of the kth bus; B,k,t Let be the station-state variable of the k-th bus at time t; The maximum charging and discharging power of the k-th bus; The number of charging parking spaces at bus stops; The maximum charging and discharging power of the bus station is denoted by ; min represents the minimum value function.

5. The electric vehicle cluster collaborative scheduling method for multi-scenario vehicle-to-grid interaction according to claim 1, characterized in that, The adjustable capability modeling for various typical electric vehicle charging scenarios is performed, and the maximum charging / discharging power and maximum adjustable energy for each scenario are calculated, including: For the aforementioned urban distribution logistics scenario, a Poisson distribution is used to simulate the number of delivery batches arriving per unit time, and a log-normal distribution is used to simulate the number of logistics vehicles arriving in each batch. The initial state of charge of each logistics vehicle is simulated using a uniform distribution. Based on the number of logistics vehicles, the initial state of charge of each vehicle, battery capacity parameters, and charging facility capacity parameters, the maximum adjustable energy and maximum charge / discharge power are calculated using the following formulas: ； In the formula, Q L,t This represents the maximum adjustable energy for the urban distribution logistics scenario. P L,t This refers to the maximum charging and discharging power in the urban distribution logistics scenario. For the first m The initial state of charge of the logistics vehicle; η L,t For a moment t The number of logistics vehicles arriving in each batch; Φ represents the minimum battery capacity of the logistics vehicles. L,m,t For the first m The state variables of a logistics vehicle; C L,m For the first m Battery capacity of a logistics vehicle; For the first m The maximum charging and discharging power of the logistics vehicle; The number of charging stations at the logistics station; The maximum charging and discharging power of the logistics station is denoted by ; min represents the minimum value function.

6. The electric vehicle cluster collaborative scheduling method for multi-scenario vehicle-network interaction according to claim 1, characterized in that, The adjustable capability modeling for various typical electric vehicle charging scenarios is performed, and the maximum charging / discharging power and maximum adjustable energy for each scenario are calculated, including: For the heavy-duty truck battery swapping scenario, the expected number of available batteries is calculated based on a queuing theory model; The initial state of charge of each backup battery is simulated using a beta distribution. Based on the expected number of available batteries, the initial state of charge of each backup battery, battery capacity parameters, and charging facility capacity parameters, the adjustable energy and maximum charge / discharge power are calculated using the following formulas: ； In the formula, Q H,t This refers to the maximum adjustable energy for the heavy-duty truck battery swapping scenario. P H,t This refers to the maximum charging and discharging power in the heavy-duty truck battery swapping scenario. For a moment t The expected number of available batteries; This represents the initial state of charge of the nth battery. This is the lower limit of the battery capacity for heavy-duty truck battery swapping stations; Let n be the capacity of the nth battery. This refers to the number of battery swapping locations; This is the maximum power of the heavy-duty truck battery swapping station; denoted as the maximum charge / discharge power of the nth battery; min represents the minimum value function.

7. The electric vehicle cluster collaborative scheduling method for multi-scenario vehicle-to-grid interaction according to claim 1, characterized in that, The function of the overall return is as follows: ； Where max represents the maximization function; Q represents the overall return; This indicates the revenue from electricity generated by participating in the day-ahead market; This indicates the revenue from participating in the real-time market for electrical energy; This indicates the returns from participating in the uphill market; Indicates the returns from participating in the downhill market; C deg This represents the cost of battery degradation caused by charging and discharging operations in electric vehicles.

8. The electric vehicle cluster collaborative scheduling method for multi-scenario vehicle-to-grid interaction according to claim 1, characterized in that, The constraints include power constraints, regulation capacity constraints, and energy dynamic constraints. The power constraints are as follows: ； The adjustment capacity constraint is as follows: ； The energy dynamic constraints are as follows: ； In the formula, This represents the day-ahead market charging level at time t in scenario s; This represents the day-ahead market discharge volume at time t in scenario s; This represents the application volume of the climbing market per unit time at time t in scenario s; express s In the scene t The application volume of the downhill market per unit time; P s,t,max This represents the maximum charge / discharge power for scenario s; express s In the scene t Real-time market charging power; express s In the scene t Real-time market discharge volume at any given moment; This indicates the minimum application capacity for the climbing market; This indicates the minimum application capacity for the downhill market; η cha Indicates charging efficiency; η dis Indicates discharge efficiency; Δt represents a unit of time; SOC min Indicates the lowest initial state of charge; E s express t Battery energy at any moment; E s,t,max express s In the scene t The maximum adjustable energy at any given time.

9. The electric vehicle cluster collaborative scheduling method for multi-scenario vehicle-to-grid interaction according to claim 1, characterized in that, The method employs Information Gap Decision Theory (IGDT) to quantify the uncertainty of the winning probability in the market capacity ramp-up, transforming the two-stage stochastic optimization model into an IGDT robust model, including: Define an uncertainty radius; the uncertainty radius includes a first uncertainty radius parameter and a second uncertainty radius parameter; the first uncertainty radius parameter represents the relative deviation of the probability of winning a bid in an upward climbing market; the second uncertainty radius parameter represents the relative deviation of the probability of winning a bid in a downward climbing market; Set the first and second uncertain radius parameters to zero, and when the winning probability in the climbing market is the benchmark value, solve the two-stage stochastic optimization model to obtain the benchmark profit; Based on the aforementioned benchmark profit, an acceptable minimum profit is set; With the goal of maximizing the tolerance for uncertainty and the constraint that the overall return is not less than the minimum profit, an IGDT robust model is constructed.

10. The electric vehicle cluster collaborative scheduling method for multi-scenario vehicle-to-grid interaction according to claim 9, characterized in that, The robust model of IGDT is as follows: ； Where max represents the maximization function; α 1 represents the first uncertain radius parameter; α 2 represents the second uncertain radius parameter; This indicates the probability of winning a bid in the uphill market; This indicates the probability of winning a bid in the downhill market. γ This represents the profit loss rate; z This represents the benchmark profit; Q This represents the overall benefit; x This represents the set of decision variables in the current market. y This represents the set of decision variables in a real-time market.

Images