An electric vehicle joint rebalancing and vehicle network interaction collaborative optimization method, system and medium

Abstract

The application provides an electric vehicle joint rebalancing and vehicle network interaction collaborative optimization method, system and medium, and the method comprises the following specific steps: S1: adopting an utility function to model the discharging utility of an electric vehicle driver; S2: proposing a joint rebalancing and vehicle network interaction collaborative game model of the electric vehicle; S3: designing a semi-distributed algorithm to solve an equilibrium point, and realizing vehicle network interaction collaborative optimization. The application synchronously realizes vehicle rebalancing and V2G discharging collaboration, protects privacy through distributed calculation, quickly converges to an equilibrium, and significantly improves scheduling efficiency and power grid support capacity.

Description

Technical Field

[0001] This application relates to the field of demand-side management, specifically a method, system, and medium for joint rebalancing of electric vehicles and collaborative optimization of vehicle-to-grid interaction. Background Technology

[0002] Shared electric vehicles (EVs) have significantly improved the convenience of travel by providing on-demand mobility services. However, due to the uneven distribution of travel demand, there is a significant discrepancy between the availability of EVs and people's travel needs, thus necessitating effective rebalancing and scheduling of EVs. The rebalancing problem requires consideration of the interaction between EVs and the power grid. Existing technologies primarily consider the interaction between EVs and the power grid, but these are all based on the charging behavior of EVs, neglecting their ability to feed electrical energy back to the grid through vehicle-to-grid (V2G) interaction technologies. Furthermore, EV research relies on complete driver information; however, in real-world scenarios, obtaining complete driver preference information is often difficult and may overlook driver privacy protection needs. Therefore, further exploration of collaborative optimization methods for EV driver rebalancing and V2G interaction is required. Summary of the Invention

[0003] The purpose of this application is to provide a method, system, and medium for joint rebalancing of electric vehicles and coordinated optimization of vehicle-grid interaction, which significantly improves scheduling efficiency and grid support capabilities.

[0004] To achieve the above objectives, this application provides the following technical solution:

[0005] In a first aspect, embodiments of this application provide a method for joint rebalancing and vehicle-to-grid interaction collaborative optimization of electric vehicles, comprising the following specific steps:

[0006] S1: The discharge utility of electric vehicle drivers is modeled using utility functions;

[0007] S2: Proposes a joint rebalancing and vehicle-network interaction collaborative aggregation game model for electric vehicles;

[0008] S3: Design a semi-distributed algorithm to solve for the equilibrium point and realize vehicle-network interaction and collaborative optimization.

[0009] In S1, we first consider a transportation network containing N electric vehicles and S rebalancing points. Charging stations, as connecting nodes of the coupled system, are equipped with bidirectional chargers to provide vehicle-to-grid interaction services. Each electric vehicle driver maximizes their own benefits by choosing a destination and a route, as detailed below:

[0010] Transportation networks use directed graphs G T(V,E) represents the set of vertices in the network, where a subset Vs⊆V represents a charging station; E is the set of road links connecting the vertices; and N represents the set of electric vehicle queues, with each electric vehicle represented by a triplet. Indicates; among which, This represents the probability of choosing path link e; This represents the probability of choosing either a charging station or a rebalancing point s; Indicates at destination p i The discharge power, d i ≥0; Let be the upper limit of the maximum discharge power of the i-th electric vehicle.

[0011] Charging stations: Charging stations are located at the vertices of the transportation network and connected to the nodes of the power network. The electricity purchase price for each charging station is known.

[0012] Traffic control center: Responsible for coordinating electric vehicles, which involves collecting route and destination selection information from all drivers and broadcasting this information to each driver.

[0013] Secondly, driver utility should be considered from the following four aspects:

[0014] 1. Discharge effect

[0015] Discharge utility refers to the overall benefit and satisfaction gained by electric vehicle drivers when feeding electrical energy back to the grid through vehicle-to-grid (V2G) interaction technology. It directly influences whether and to what extent drivers are willing to discharge energy back to the grid. The satisfaction gained by driver i∈N through providing V2G interaction services is represented by the utility function. ,in Indicates charging station s∈V cs The discharge electricity price; p s The probability component for the driver to choose a charging station or rebalancing point s;

[0016] Property 1: Discharge efficiency varies with discharge power d i Non-decreasing, meaning drivers tend to feed electrical energy back into the grid, can be expressed mathematically as:

[0017] (2)

[0018] Property 2: Discharge effect on d i The second derivative is not positive, that is:

[0019] (3)

[0020] Property 3: Discharge utility is related to electricity price; the higher the electricity price, the greater the utility gained by the driver.

[0021] (4)

[0022] Property 4: The discharge utility exists only when providing vehicle-to-grid (V2G) interaction services, that is:

[0023] (5)

[0024] Based on the above properties, the quadratic discharge utility function is designed as follows:

[0025] (6)

[0026] in This is the first type of parameter specific to driver i.

[0027] 2. Route and destination preferences

[0028] Drivers may have preferences for routes and destinations based on past experience. Let driver i's preferred routes and destinations be represented as follows:

[0029] (7)

[0030] This represents the "preferred route vector" for the i-th driver. For vectors The e-th component in the equation represents the "preference probability of the i-th driver for the e-th road link". This indicates that the component values ​​range from 0 to 1; Let S be the "preferred destination vector" of the i-th driver, and S be the total number of rebalancing points; Let be the probability of the i-th driver's preference for the s-th rebalancing point.

[0031] Then driver i deviates from the preferred route and destination The utility loss is described by a quadratic function:

[0032] (8)

[0033] in This is the second type of parameter specific to driver i.

[0034] 3. Travel costs

[0035] The travel time of a road link is an increasing function of traffic flow, as shown in the following formula:

[0036] (9)

[0037] t e The travel time for road segment e is... and c e These represent the free-flow travel time of link e and the "actual link capacity," respectively; and the link traffic. The expression for the total expected number of electric vehicles is:

[0038] (10)

[0039] in Let be the number of electric vehicles on link e. Let m be the link traffic parameter, used to quantify the correspondence between link traffic and the number of electric vehicles.

[0040] The expected travel utility for driver i∈N is:

[0041] (11)

[0042] , Let E be the travel cost coefficient for driver i, and E be the set of travel segments.

[0043] 4. Destination rebalancing utility

[0044] To avoid excessive vehicle concentration at certain destinations, a congestion rebalancing effect proportional to the number of electric vehicles clustered at those destinations is introduced:

[0045] (12)

[0046] To assess the adequacy of drivers under future travel demand, The larger the value, the lower the rebalancing utility. The sum of the proportions of all drivers choosing this point at destination s is expressed as:

[0047] (13)

[0048] The rebalancing utility of driver i is: (14)

[0049] For rebalancing cost coefficients,

[0050] 5. Total utility function

[0051] The total utility of driver i consists of discharge utility Route destination preference utility Travel cost-utility and destination congestion rebalancing effect The composition is as follows: The first two items are only affected by the individual driver's choice, while the latter two items depend not only on local variables but also on the average value of the overall strategy, i.e.:

[0052] (15)

[0053] definition:

[0054] (16)

[0055] (17)

[0056] The total utility of driver i can then be expressed as: (18)

[0057] The avg(.) function outputs the average of a set of data corresponding to the variable. i It concerns the independent variable strategy x i A function of the average value avg(x); The utility portion is determined solely by the driver's own choice; The utility component affected by the overall behavior of other drivers;

[0058] 6. Transportation network constraints

[0059] (1) Flow continuity constraint ,

[0060] Drivers must ensure traffic flow continuity when driving:

[0061] (19)

[0062]

[0063] e:(j,k) represents a standard road segment, where j is the starting point and k is the ending point;

[0064] Where driver i starts from the initial position Starting with probability Arrive at the destination Where V is the set of all drivers, V cs For a collection of all possible destinations or charging stations;

[0065] (2) Road capacity constraint C e

[0066] Traffic flow on a road link must not exceed the road capacity:

[0067] (20)

[0068] This refers to the maximum road capacity cost for road segment e.

[0069] (3) Charging station capacity constraint C cs

[0070] The number of electric vehicles that can be charged at a charging station is limited by the number of charging piles.

[0071] (twenty one)

[0072] The number of charging piles corresponding to the destination or charging station s;

[0073] The coupling constraint is then:

[0074] (twenty two)

[0075] The model's set of coupling constraints C is the road segment constraint set C. e With charging station constraint set C cs The intersection of.

[0076] In S2, each driver is first defined by selecting strategy x. i To maximize their own utility, their choices are influenced by the aggregation behavior of other drivers, while also adhering to traffic network capacity constraints and electricity price constraints.

[0077] Then comes model building:

[0078] The electric vehicle driver aggregation game G with coupling constraints is defined as follows:

[0079] (twenty three)

[0080] G represents the entire game model, and N represents the set of participants in the game. It is the set of utility functions for each driver i. It is the set of policy spaces for each driver i.

[0081] The feasible region of the game is:

[0082] (twenty four)

[0083] X represents the feasible joint strategy space of the entire system, which is the set of strategy combinations that all drivers can choose simultaneously, satisfying all constraints. This is the Cartesian product of the individual strategy spaces of all drivers. The intersection operation means that only policy combinations that satisfy all coupling constraints are retained.

[0084] The joint strategy set of all drivers is (25)

[0085] If a joint strategy If the following conditions are met, it is called a generalized clustering equilibrium of game G:

[0086] (26)

[0087] (27)

[0088] This means that when the aggregation strategies of other drivers are fixed, any driver i cannot achieve higher efficiency by choosing other strategies.

[0089] In S3, a semi-distributed algorithm is used to find the equilibrium state. A semi-distributed algorithm means that it does not require a central controller to collect information from all drivers. Instead, drivers make decisions locally and exchange only a small amount of information with their neighbors or the system, thus converging to an equilibrium state.

[0090] First, we introduce a pseudo gradient mapping.

[0091] This refers to the mapping from the policy space X to the N-dimensional real space, representing the gradient vector of the entire system.

[0092] (28)

[0093] U is the utility function of driver i i Regarding its own strategy x i The gradient; this mapping consists of the gradient of each driver's utility function with respect to its own policy. The driver's utility function does not involve the discharge utility with coupling constraints. and route destination Preference utility is a convex function, while travel costs and rebalancing costs, which involve coupling constraints, are linear functions. An inertial forward-reflection-backward algorithm is used to solve for the variational generalized clustering equilibrium point.

[0094] The algorithm is as follows:

[0095] After initialization, let (29)

[0096] in The momentum coefficient, For stability parameters, Lipschitz constant

[0097] (30)

[0098] Let i be the initial strategy for driver i; This is a pre-initialized strategy; both are n-dimensional real vectors.

[0099] (31)

[0100] (32)

[0101] For local step size, For global step size, Let A be the constraint matrix. i norm, λ is the dual variable. 0 Let λ be the initial value of the dual variable; 1 The initial values ​​for the dual variable λ are given.

[0102] Then iterate until convergence:

[0103] For all :

[0104] (30)

[0105] in, The gradient correction term for driver i at the k-th iteration.

[0106] (31)

[0107] (32)

[0108] For proximal operators, The general constraint matrix represents the linear relationship between driver i's strategy and the global coupling constraints. For the transpose of the constraint matrix, b i To constrain the target value, and These represent the weight coefficients of the constraint function;

[0109] The dual variable is updated as follows:

[0110] (33)

[0111] It is a non-negative projection operator.

[0112] Secondly, embodiments of this application provide an electric vehicle joint rebalancing and vehicle-to-grid interaction collaborative optimization system. The system includes a memory and a processor. The memory includes a program for an electric vehicle joint rebalancing and vehicle-to-grid interaction collaborative optimization method. When the program for the electric vehicle joint rebalancing and vehicle-to-grid interaction collaborative optimization method is executed by the processor, it implements the steps described above.

[0113] Thirdly, embodiments of this application provide a computer-readable storage medium storing program code, which, when executed by a processor, implements the steps of the electric vehicle joint rebalancing and vehicle-to-grid interaction collaborative optimization method as described above.

[0114] Compared with the prior art, the beneficial effects of the present invention are: compared with the prior art that only considers charging and relies on centralized control, it can simultaneously realize vehicle rebalancing and V2G discharge coordination, distributed computing protects privacy, quickly converges to equilibrium, and significantly improves scheduling efficiency and grid support capabilities. Attached Figure Description

[0115] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded 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.

[0116] Figure 1 A flowchart of a method for joint rebalancing and vehicle-to-grid interaction collaborative optimization of electric vehicles provided in this disclosure embodiment;

[0117] Figure 2 This is a diagram of the Nguyen-Dupuy traffic network structure provided in the embodiments of this disclosure;

[0118] Figure 3 This is a diagram of the IEEE 33-node power system network structure provided in the embodiments of this disclosure;

[0119] Figure 4 This diagram illustrates the power demand of each node and the corresponding electricity price of each charging station during vehicle-to-grid (V2G) interaction and vehicle-to-grid (V2G) interaction as described in the embodiments of this disclosure.

[0120] Figure 5 This is a map showing the electric vehicle arrival rate of each charging station provided in the embodiments of this disclosure. Detailed Implementation

[0121] The technical solutions of the embodiments of this application will now be described with reference to the accompanying drawings. It should be noted that similar reference numerals and letters in the following drawings indicate similar items; therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0122] The terms “comprising,” “including,” or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus 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 apparatus. Without further limitation, an element defined by the phrase “comprising one…” does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0123] The terms “first,” “second,” etc., are used only to distinguish one entity or operation from another, and should not be construed as indicating or implying relative importance, nor as requiring or implying any such actual relationship or order between these entities or operations.

[0124] This disclosure provides a method for optimizing vehicle-to-network interaction based on master-slave game theory, including:

[0125] S1: The discharge utility of electric vehicle drivers is modeled using utility functions;

[0126] S2: Proposes a joint rebalancing and vehicle-network interaction collaborative aggregation game model for electric vehicles;

[0127] S3: Design a semi-distributed algorithm to find the equilibrium point;

[0128] S4: The method has a significant rebalancing effect, as verified by numerical examples.

[0129] In S1, we first consider a transportation network comprising N electric vehicles and S rebalancing points (including charging stations). Charging stations, acting as connecting nodes in the coupled system, are equipped with bidirectional chargers to provide vehicle-to-grid (V2G) interaction services. Each electric vehicle driver maximizes their own benefits by selecting their destination and route. A detailed description follows:

[0130] Transportation networks use directed graphs G T (V,E) represents a network where V is the set of vertices, and its subset Vs⊆V represents charging stations; E is the set of road links connecting the vertices. N represents the set of electric vehicle queues. Each electric vehicle is represented by a triplet. Indicates; among which, This represents the probability of choosing path link e; This represents the probability of choosing either a charging station or a rebalancing point s; Indicates at destination p i The discharge power. d i ≥0; The upper limit of the maximum discharge power of the i-th electric vehicle

[0131] Charging stations: Charging stations are located at the top of the transportation network and connected to the nodes of the power network. The electricity purchase price for each charging station is known.

[0132] Traffic Control Center: Responsible for coordinating electric vehicles, which involves collecting route and destination selection information from all drivers and broadcasting the relevant information back to each driver.

[0133] Secondly, driver utility should be considered from the following four aspects:

[0134] 1. Discharge effect

[0135] Discharge utility refers to the overall benefit and satisfaction gained by electric vehicle drivers when feeding electrical energy back to the grid through vehicle-to-grid (V2G) interaction technology. It directly influences whether and to what extent drivers are willing to discharge energy back to the grid. The satisfaction gained by driver i∈N through providing V2G interaction services is represented by the utility function. ,in Indicates charging station s∈V cs The discharge electricity price; p s The probability component for the driver to choose a charging station or rebalancing point s;

[0136] Property 1: Discharge efficiency varies with discharge power d i Non-decreasing, meaning drivers tend to feed electrical energy back into the grid, can be expressed mathematically as:

[0137] (2)

[0138] Property 2: Discharge effect on d i The second derivative is not positive, that is:

[0139] (3)

[0140] Property 3: Discharge utility is related to electricity price; the higher the electricity price, the greater the utility gained by the driver.

[0141] (4)

[0142] Property 4: The discharge utility exists only when providing vehicle-to-grid (V2G) interaction services, that is:

[0143] (5)

[0144] Based on the above properties, the quadratic discharge utility function is designed as follows:

[0145] (6)

[0146] Where β 1,i ,β 2,i ≥0 is the first type of exclusive parameter for driver i.

[0147] 2. Route and destination preferences

[0148] Drivers may have preferences for routes and destinations based on past experience. Let driver i's preferred routes and destinations be represented as follows:

[0149] (7)

[0150] This represents the "preferred route vector" for the i-th driver. For vectors The e-th component in the equation represents the "preference probability of the i-th driver for the e-th road link". This indicates that the component values ​​range from 0 to 1; Let S be the "preferred destination vector" for the i-th driver, and S be the total number of rebalancing points (charging stations). Let be the probability of the i-th driver's preference for the s-th rebalancing point (charging station).

[0151] Then driver i deviates from the preferred route and destination The utility loss is described by a quadratic function:

[0152] (8)

[0153] Where α 1,i α 2,i ≥0 is the second type of exclusive parameter for driver i.

[0154] 3. Travel costs

[0155] The travel time of a road link is an increasing function of traffic flow, as shown in the following formula:

[0156] (9)

[0157] t e The travel time for road segment e is... and c e These represent the free-flow travel time of link e and the "actual link capacity," respectively; the link traffic ϕ. e (v e Let be a function of the total expected number of electric vehicles, expressed as:

[0158] (10)

[0159] in Let be the number of electric vehicles on link e. m is the link traffic parameter, used to quantify the correspondence between link traffic and the number of electric vehicles.

[0160] The expected travel utility for driver i∈N is:

[0161] (11)

[0162] , Let E be the travel cost coefficient for driver i, and E be the set of travel segments.

[0163] 5. Destination rebalancing utility

[0164] To avoid excessive vehicle concentration at certain destinations, a congestion rebalancing effect proportional to the number of electric vehicles clustered at those destinations is introduced:

[0165] (12)

[0166] To assess the adequacy of drivers under future travel demand, The larger the value, the lower the rebalancing utility. The sum of the proportions of all drivers choosing this point at destination s is expressed as:

[0167] (13)

[0168] The rebalancing utility of driver i is: (14)

[0169] This is the rebalancing cost coefficient.

[0170] 5. Total utility function

[0171] The total utility of driver i consists of discharge utility Route destination preference utility Travel cost-utility and destination congestion rebalancing effect The composition is as follows: The first two items are only affected by the individual driver's choice, while the latter two items depend not only on local variables but also on the average value of the overall strategy, i.e.:

[0172] (15)

[0173] definition:

[0174] (16)

[0175] (17)

[0176] The total utility of driver i can then be expressed as: (18)

[0177] The `avg(.)` function outputs the average of a set of data corresponding to a variable. i It concerns the independent variable strategy x i A function of the average value avg(x); The utility portion is determined solely by the driver's own choice; The utility component affected by the overall behavior of other drivers;

[0178] 6. Transportation network constraints:

[0179] (1) Flow continuity constraint ,

[0180] Drivers must ensure traffic flow continuity when driving:

[0181] (19)

[0182]

[0183] e:(j,k) represents a standard road segment, where j is the starting point and k is the ending point;

[0184] Where driver i starts from the initial position Starting with probability Arrive at the destination Where V is the set of all drivers, V cs A collection of all available destinations or charging stations.

[0185] (4) Road capacity constraint C e

[0186] Traffic flow on a road link must not exceed the road capacity:

[0187] (20)

[0188] The road capacity cap cost for road segment e

[0189] (5) Charging station capacity constraints ,

[0190] The number of electric vehicles that can be charged at a charging station is limited by the number of charging piles.

[0191] (twenty one)

[0192] The number of charging piles corresponding to the destination or charging station s;

[0193] The coupling constraint is then:

[0194] (twenty two)

[0195] The model's set of coupling constraints C is the road segment constraint set C. e With charging station constraint set C cs The intersection;

[0196] In S2, each driver is first defined by selecting strategy x. i To maximize its own utility, its choices are influenced by the aggregation behavior of other drivers, while also adhering to traffic network capacity constraints and electricity price constraints.

[0197] Then comes model building:

[0198] The electric vehicle driver aggregation game G with coupling constraints is defined as follows:

[0199] (twenty three)

[0200] G represents the entire game model, and N represents the set of participants in the game. It is the set of utility functions for each driver i. It is the set of policy spaces for each driver i.

[0201] The feasible region of the game is:

[0202] (twenty four)

[0203] X represents the feasible joint strategy space of the entire system, which is the set of strategy combinations that all drivers can choose simultaneously and that satisfy all constraints. This is the Cartesian product of the individual strategy spaces of all drivers. The intersection operation means that only policy combinations that satisfy all coupling constraints are retained.

[0204] The joint strategy set of all drivers is (25)

[0205] If a joint strategy If the following conditions are met, it is called a generalized clustering equilibrium of game G:

[0206] (26)

[0207] (27)

[0208] This means that when the aggregation strategies of other drivers are fixed, any driver i cannot achieve higher efficiency by choosing other strategies.

[0209] In S3, a semi-distributed algorithm is used to solve for the equilibrium. A semi-distributed algorithm means that it does not require a central controller to collect information from all drivers. Instead, drivers make decisions locally and exchange only a small amount of information with their neighbors or the system, so that they can converge to an equilibrium state.

[0210] First, we introduce a pseudo gradient mapping.

[0211] This refers to the mapping from the policy space X to the N-dimensional real space, representing the gradient vector of the entire system.

[0212] (28)

[0213] U is the utility function of driver i i Regarding its own strategy x i The gradient; this mapping consists of the gradient of each driver's utility function with respect to its own policy. The driver's utility function does not involve the discharge utility with coupling constraints. and route destination Preference utility is a convex function, while travel costs and rebalancing costs, which involve coupling constraints, are linear functions. An inertial forward-reflection-backward algorithm is used to solve for the variational generalized clustering equilibrium point.

[0214] The algorithm is as follows:

[0215] After initialization, let (29)

[0216] in The momentum coefficient, For stability parameters, Lipschitz constant

[0217] (30)

[0218] Let i be the initial strategy for driver i; This is a pre-initialized strategy; both are n-dimensional real vectors.

[0219] (31)

[0220] (32)

[0221] For local step size, For global step size, Let A be the constraint matrix. i norm, λ is the dual variable. 0 Let λ be the initial value of the dual variable; 1λ is the initial value of the dual variable λ.

[0222] Then iterate until convergence:

[0223] For all :

[0224] (30)

[0225] in, The gradient correction term for driver i at the k-th iteration.

[0226] (31)

[0227] (32)

[0228] For proximal operators, is the general constraint matrix, representing the linear relationship between driver i's strategy and the global coupling constraints. For the transpose of the constraint matrix, b i To constrain the target value, and These represent the weight coefficients of the constraint function;

[0229] The dual variable is updated as follows:

[0230] (33)

[0231] It is a non-negative projection operator;

[0232] In S4, a coupled system was constructed based on the Nguyen-Dupui transportation network with 13 nodes and 19 links and the IEEE 33-node power system to verify the feasibility of the proposed model. The network structure is as follows: Figure 2 and Figure 3 As shown, the two networks are interconnected via charging stations (labeled "CS1-CS5" in the diagram). The baseline values ​​for traffic flow and electricity demand are set at 10,000 vehicles / hour and 10 MVA, respectively, with the maximum discharge power of electric vehicles being 300 watts. Five nodes in the distribution network supply power to the charging stations and correspond to five vertices in the traffic network. The coupling relationships are as follows: CS1 matches node 4, CS2 matches node 9, CS3 matches node 3, CS4 matches node 13, and CS5 matches node 2.

[0233] Electricity price T at each node i This is determined by introducing an importance indicator:

[0234] (34)

[0235] (35)

[0236] Where 100 is the conversion factor; m is the global baseline factor, used to map the abstract cost value inside the model to the actual currency unit; , Unique load attributes that respectively measure the importance of load to life safety and socio-economic factors; and The value range is {1, 2, 3, 4, 5}. The electricity purchase prices for CS1-CS5 are set at 72, 60, 36, 24 and 48 yuan / MWh, respectively.

[0237] Figure 4 This study compared the power demand at five charging stations with and without vehicle-to-everything (V2X) services. The left diagonal bar chart represents the power demand when no electric vehicles are discharging, while the right diagonal bar chart represents the power demand when electric vehicles are discharging. The difference between the two charts represents the discharge power of electric vehicles, and the dashed lines represent the electricity price at each charging station. The results show that all charging stations have electric vehicle discharge power, which can alleviate power shortages in scenarios of power congestion. However, due to differences in driver preferences and electricity purchase prices at various charging stations, the distribution of electric vehicle selection at different charging stations is uneven.

[0238] Figure 5 The convergence process of electric vehicle arrival rates at various charging stations is illustrated. Due to differences in electricity prices at each charging station, the discharge flow of electric vehicles to each station varies, resulting in different convergence values ​​for the arrival rates. The algorithm nearly converges on the 16th iteration and fully converges on the 41st iteration.

[0239] Depend on Figure 4 and Figure 5The coupling results show that under the proposed semi-distributed game theory method, the system exhibits a typical equilibrium characteristic: "charging stations with low discharge demand have high EV arrival rates, while charging stations with low discharge prices have low EV arrival rates." On the one hand, stations with lower discharge demand, due to less grid discharge regulation pressure and relatively abundant resources, accept higher EV arrival rates under the guidance of game coordination signals, realizing the transfer of discharge resources to low-demand areas and alleviating the regulation load on high-discharge-demand stations. On the other hand, stations with lower discharge prices do not experience disorderly EV influx, and their arrival rates remain within a reasonable range. This indicates that the dynamic discharge price, with dual variables as its core, can accurately reflect the degree of constraint on stations, guiding users to make rational charging and discharging participation decisions and avoiding local station resource congestion or regulation redundancy. These characteristics collectively demonstrate that the proposed game theory method can achieve a balanced distribution of electric vehicles among charging stations through price incentives and distributed strategy iteration, completing a precise spatial rebalancing of discharge resources and grid demand in the vehicle-grid interaction scenario, and enabling the system to reach a Nash equilibrium state that unifies individual rationality and global balance, fully verifying the significant rebalancing effect and coordination effectiveness of the method.

[0240] This application provides an electric vehicle joint rebalancing and vehicle-to-grid interaction collaborative optimization system. The system includes a memory and a processor. The memory includes a program for an electric vehicle joint rebalancing and vehicle-to-grid interaction collaborative optimization method. When the program for the electric vehicle joint rebalancing and vehicle-to-grid interaction collaborative optimization method is executed by the processor, it implements the steps described above.

[0241] This application provides a computer-readable storage medium storing program code. When the program code is executed by a processor, it implements the steps of the electric vehicle joint rebalancing and vehicle-to-grid interaction collaborative optimization method as described above.

[0242] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0243] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0244] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0245] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0246] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0247] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0248] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0249] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for joint rebalancing and vehicle-to-grid interaction collaborative optimization of electric vehicles, characterized in that, The specific steps include the following: S1: The discharge utility of electric vehicle drivers is modeled using utility functions; S2: Proposes a joint rebalancing and vehicle-network interaction collaborative aggregation game model for electric vehicles; S3: Design a semi-distributed algorithm to solve for the equilibrium point and realize vehicle-network interaction and collaborative optimization.

2. The method for joint rebalancing and vehicle-network interaction collaborative optimization of electric vehicles according to claim 1, characterized in that, In S1, we first consider a transportation network containing N electric vehicles and S rebalancing points. Charging stations, as connecting nodes of the coupled system, are equipped with bidirectional chargers to provide vehicle-to-grid interaction services. Each electric vehicle driver maximizes their own benefits by choosing a destination and a route, as detailed below: Transportation networks use directed graphs G T (V,E) represents the set of vertices in the network, where a subset Vs⊆V represents a charging station; E is the set of road links connecting the vertices; and N represents the set of electric vehicle queues, with each electric vehicle represented by a triplet. express; Among them, v i =[ ] represents the probability of choosing path link e; p i =[ ] s∈V d represents the probability of choosing either the charging station or the rebalancing point s; i ∈[0, ] indicates at destination p i The discharge power, d i ≥0; Let be the upper limit of the maximum discharge power of the i-th electric vehicle. Charging stations: Charging stations are located at the vertices of the transportation network and connected to the nodes of the power network. The electricity purchase price for each charging station is known. Traffic control center: Responsible for coordinating electric vehicles, which involves collecting route and destination selection information from all drivers and broadcasting this information to each driver. Secondly, driver utility should be considered from the following four aspects: Discharge effect: Discharge utility refers to the overall benefit and satisfaction gained by electric vehicle drivers when feeding electrical energy back to the grid through vehicle-to-grid (V2G) interaction technology. It directly influences whether and to what extent drivers are willing to discharge energy back to the grid. The satisfaction gained by driver i∈N through providing V2G interaction services is represented by the utility function. ,in Indicates charging station s∈V cs The discharge electricity price; p s The probability component for the driver to choose a charging station or rebalancing point s; Property 1: Discharge efficiency varies with discharge power d i Non-decreasing, meaning drivers tend to feed electrical energy back into the grid, can be expressed mathematically as: （2） Property 2: Discharge effect on d i The second derivative is not positive, that is: （3） Property 3: Discharge utility is related to electricity price; the higher the electricity price, the greater the utility gained by the driver. （4） Property 4: The discharge utility exists only when providing vehicle-to-grid (V2G) interaction services, that is: （5） Based on the above properties, the quadratic discharge utility function is designed as follows: （6） Where β 1,i ,β 2,i ≥0 is a first-type specific parameter for driver i. Route and destination preferences: Drivers may have preferences for routes and destinations based on past experience. Let driver i's preferred routes and destinations be represented as follows: （7） This represents the "preferred route vector" of the i-th driver. For vectors The e-th component in the equation represents the "preference probability of the i-th driver for the e-th road link". This indicates that the component values ​​range from 0 to 1; Let S be the "preferred destination vector" of the i-th driver, and S be the total number of rebalancing points; Let be the probability of the i-th driver's preference for the s-th rebalancing point. Then driver i deviates from the preferred route and destination The utility loss is described by a quadratic function: （8） Where α 1,i α 2,i ≥0 is a type II specific parameter for driver i. Travel costs: The travel time of a road link is an increasing function of traffic flow, as shown in the following formula: （9） t e The travel time for road segment e is... and c e These represent the free-flow travel time and "actual link capacity" of link e, respectively; and the link traffic ϕ. e (v e Let be a function of the total expected number of electric vehicles, expressed as: （10） in Let be the number of electric vehicles on link e. Let m be the link traffic parameter, used to quantify the correspondence between link traffic and the number of electric vehicles. The expected travel utility for driver i∈N is: （11） , Let E be the travel cost coefficient for driver i, and E be the set of travel segments. Destination rebalancing utility: To avoid excessive vehicle concentration at certain destinations, a congestion rebalancing effect proportional to the number of electric vehicles clustered at those destinations is introduced: （12） To assess the adequacy of drivers under future travel demand, The larger the value, the lower the rebalancing utility. The sum of the proportions of all drivers choosing this point at destination s is expressed as: （13） The rebalancing utility of driver i is: (14) For rebalancing cost coefficients, Total utility function: The total utility of driver i consists of discharge utility Route destination preference utility Travel cost-utility and destination congestion rebalancing effect The composition is as follows: The first two items are only affected by the individual driver's choice, while the latter two items depend not only on local variables but also on the average value of the overall strategy, i.e.: （15） definition: （16） （17） Then the total utility of driver i can be expressed as: U i (x i ,avg(x))=g i +f i (18) The avg(.) function outputs the average of a set of data corresponding to the variable. i It concerns the independent variable strategy x i A function of the average value avg(x); The utility portion is determined solely by the driver's own choice; The utility component affected by the overall behavior of other drivers; Traffic network constraints: (1) Flow continuity constraint , Drivers must ensure traffic flow continuity when driving: （19） ，， e:(j,k) represents a standard road segment, where j is the starting point and k is the ending point; Where driver i starts from the initial position Starting with probability Arrive at the destination Where V is the set of all drivers, V cs For a collection of all possible destinations or charging stations; (2) Road capacity constraint C e Traffic flow on a road link must not exceed the road capacity: (20) This refers to the maximum road capacity cost for road segment e. (3) Charging station capacity constraint C cs The number of electric vehicles that can be charged at a charging station is limited by the number of charging piles. (21) The number of charging piles corresponding to the destination or charging station s; The coupling constraint is then: (22) The model's set of coupling constraints C is the road segment constraint set C. e With charging station constraint set C cs The intersection of.

3. The method for joint rebalancing and vehicle-network interaction collaborative optimization of electric vehicles according to claim 1, characterized in that, In S2, each driver is first defined by selecting strategy x. i To maximize their own utility, their choices are influenced by the aggregation behavior of other drivers, while also adhering to traffic network capacity constraints and electricity price constraints. Then comes model building: The electric vehicle driver aggregation game G with coupling constraints is defined as follows: (23) G represents the entire game model, and N represents the set of participants in the game. It is the set of utility functions for each driver i. It is the set of policy spaces for each driver i. The feasible region of the game is: (24) X represents the feasible joint strategy space of the entire system, which is the set of strategy combinations that all drivers can choose simultaneously, satisfying all constraints. This is the Cartesian product of the individual strategy spaces of all drivers. The intersection operation means that only policy combinations that satisfy all coupling constraints are retained. The joint strategy set of all drivers is (25) If a joint strategy If the following conditions are met, it is called a generalized clustering equilibrium of game G: （26） （27） This means that when the aggregation strategies of other drivers are fixed, any driver i cannot achieve higher efficiency by choosing other strategies.

4. The method for joint rebalancing and vehicle-network interaction collaborative optimization of electric vehicles according to claim 1, characterized in that, In S3, a semi-distributed algorithm is used to find the equilibrium state. A semi-distributed algorithm means that it does not require a central controller to collect information from all drivers. Instead, drivers make decisions locally and exchange only a small amount of information with their neighbors or the system, thus converging to an equilibrium state. First, we introduce a pseudo gradient mapping. , This refers to the mapping from the policy space X to the N-dimensional real space, representing the gradient vector of the entire system. （28） U is the utility function of driver i i Regarding its own strategy x i The gradient; this mapping consists of the gradient of each driver's utility function with respect to its own policy. The driver's utility function does not involve the discharge utility with coupling constraints. and route destination Preference utility is a convex function, while travel costs and rebalancing costs, which involve coupling constraints, are linear functions. An inertial forward-reflection-backward algorithm is used to solve for the variational generalized clustering equilibrium point. The algorithm is as follows: After initialization, let (29) in The momentum coefficient, For stability parameters, Lipschitz constant （30) Let i be the initial strategy for driver i; For pre-initial strategy; Both are n-dimensional real vectors; （31） (32) For local step size, For global step size, Let A be the constraint matrix. i norm, λ is the dual variable. 0 Let λ be the initial value of the dual variable; 1 The initial values ​​for the dual variable λ are given. Then iterate until convergence: For all : （30） in, The gradient correction term for driver i at the k-th iteration. （31) (32) For proximal operators, The general constraint matrix represents the linear relationship between driver i's strategy and the global coupling constraints. For the transpose of the constraint matrix, b i To constrain the target value, and These represent the weight coefficients of the constraint function; The dual variable is updated as follows: (33) It is a non-negative projection operator.

5. A joint rebalancing and vehicle-to-grid interaction collaborative optimization system for electric vehicles, characterized in that, The system includes a memory and a processor. The memory includes a program for a method of joint rebalancing of electric vehicles and coordinated optimization of vehicle-to-grid interaction. When the program of the method of joint rebalancing of electric vehicles and coordinated optimization of vehicle-to-grid interaction is executed by the processor, it implements the steps described above.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program code, which, when executed by a processor, implements the steps of the electric vehicle joint rebalancing and vehicle-to-grid interaction collaborative optimization method as described in any one of claims 1 to 4.

Images