Real-time charging reservation method for electric vehicle

Abstract

The invention belongs to the technical field of electric vehicle charging path planning, and particularly relates to an electric vehicle real-time charging reservation method, which comprises the following steps: S0, acquiring real-time traffic network data, the running state of each charging station and power grid load information; s1, acquiring initial state information of the electric vehicle initiating the charging request; s2, constructing an optimization model with the purpose of minimizing the comprehensive cost; s3, solving a charging station with the lowest comprehensive cost and a corresponding navigation path, and performing charging reservation; s4, when any node in the traffic network is reached, the optimization and solving operation in the S3 is executed repeatedly; updating the charging appointment; and S5, when arriving at the finally selected charging station after dynamic optimization, executing a charging service according to the latest reservation information. According to the method, multi-dimensional real-time information is fused, dynamic path re-planning and intelligent queue jumping judgment are supported, and an electric vehicle real-time charging reservation mechanism with the minimum user comprehensive travel cost is achieved on the premise that safety and traffic constraints are met.

Description

Technical Field

[0001] This invention belongs to the field of electric vehicle charging route planning technology, and particularly relates to a method for real-time charging reservation for electric vehicles. Background Technology

[0002] With the deepening of the global "dual-carbon" strategy, the electric vehicle (EV) industry has experienced explosive growth. However, the large-scale, unregulated integration of electric vehicles into urban transportation networks and power distribution systems is putting unprecedented pressure on infrastructure operations. Especially during peak electricity consumption periods or holidays, problems such as significantly longer waiting times at charging stations, increased traffic congestion on main urban roads, and overloads or even voltage exceedances at local power distribution nodes are becoming increasingly prominent, severely restricting user travel experience and the safe and stable operation of the power grid.

[0003] The current mainstream electric vehicle charging guidance and reservation strategies still have obvious limitations, mainly reflected in the following three aspects: (1) Single information dimension and delayed response: Most existing systems adopt static or quasi-static mechanisms, and only make reservation matching based on local information of "vehicle-station" or "station-network", lacking the ability to coordinate perception and rapid response to dynamic changes in traffic network (such as sudden congestion and accident detour) and real-time load fluctuations of the power grid; (2) Lack of urgency perception and priority guarantee mechanism: For high-urgency vehicles with extremely low remaining power and tight travel time, existing methods generally do not set effective queue-jumping judgment rules and priority service strategies, resulting in some users facing "range anxiety" or even the risk of breakdown in the middle of the journey; (3) Separation of route and charging decision: Traditional route planning algorithms are often optimized independently of the waiting time at charging stations, ignoring the strong coupling relationship between driving time and queuing waiting time, making the user's actual total travel cost (including time, energy consumption, opportunity cost, etc.) much higher than expected.

[0004] To address these challenges, several technical solutions have been proposed in recent years to improve the system from different perspectives. For example, patent CN202510347986.8, "An Orderly Charging Method for Electric Vehicle Charging Stations Based on Intelligent Queuing," proposes to dynamically predict the idle status of charging piles through a sliding window mechanism and construct a priority queue based on vehicle distance and waiting time to achieve efficient scheduling of resources within the station. While this method effectively improves the internal operational efficiency of charging stations, its optimization scope is limited to within the station and does not consider the path selection problem from the user's starting point to the charging station, lacking a holistic "vehicle-road-station" integrated perspective. Another example is patent CN202510192991.6, "An Intelligent Charging Path Planning Method for New Energy Vehicles Applicable to Urban Roads," which introduces an ant colony algorithm to perform path search based on a simplified road network topology, improving computational efficiency in complex urban environments. However, this solution mainly relies on static road network data and fails to integrate dynamic elements such as real-time traffic speed, flow rate, and charging station queuing status, making it difficult to adapt to sudden disturbances in real traffic scenarios.

[0005] The reason why existing technologies struggle to achieve efficient and robust charging reservation services lies in the deep coupling challenges of multi-source heterogeneous systems: on the one hand, transportation systems exhibit high spatiotemporal dynamism and uncertainty, with road capacity fluctuating dramatically over time; on the other hand, the operating status of the power distribution network is affected by multiple factors such as regional load and distributed energy output, leading to frequent changes in safety constraints; furthermore, charging station queuing behavior is nonlinear and interactive, collectively constituting a high-dimensional, non-convex, and time-varying joint optimization problem. Moreover, introducing a queue-jumping mechanism to protect the rights of users with high urgency requires a delicate balance between fairness and efficiency, along with reasonable compensation for affected users, further increasing the complexity of system design.

[0006] Therefore, how to build a real-time electric vehicle charging reservation mechanism that integrates multi-dimensional real-time information from "vehicle-road-station-network", supports dynamic route replanning and intelligent queue-jumping judgment, and minimizes the overall travel cost for users while meeting the constraints of power grid safety and traffic capacity has become an urgent problem to be solved. Summary of the Invention

[0007] To address the shortcomings of the existing technologies, this invention provides a real-time charging reservation method for electric vehicles that integrates multi-dimensional real-time information from "vehicle-road-station-network," supports dynamic route replanning and intelligent queue-jumping determination, and minimizes the overall travel cost for users while meeting the constraints of power grid safety and traffic capacity.

[0008] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0009] A method for real-time charging reservation for electric vehicles includes the following steps:

[0010] S0. Obtain real-time traffic network data, the operating status of each charging station, and power grid load information; wherein, the traffic network data includes road segment speed and traffic flow, used to construct the traffic network topology; the operating status of the charging stations includes charging pile occupancy and queue information;

[0011] S1. Obtain the initial state information of the electric vehicle that initiated the charging request; wherein, the initial state information includes the starting point, the destination, the departure time, and the remaining battery power;

[0012] S2. Based on the information obtained from S0 and S1, an optimization model is constructed with the goal of minimizing the overall cost. The optimization model introduces traffic capacity constraints, vehicle accessibility constraints, and power distribution network safety operation constraints to limit the feasible selection of charging stations and route combinations.

[0013] S3. Using the improved dynamic Floyd algorithm, based on the optimization model built in S2, calculate the corresponding comprehensive cost for each feasible charging station, and solve for the charging station with the lowest comprehensive cost and its corresponding navigation path; and make corresponding charging reservations.

[0014] In calculating the overall cost of each candidate charging station, it is simultaneously assessed whether the electric vehicle that initiated the charging request meets the conditions for queuing at that station. If it does, the queuing utility increment is included in the overall cost calculation, and corresponding compensation is provided to users affected by queuing.

[0015] S4. During the journey of the electric vehicle to the reserved charging station, whenever it reaches any node in the traffic network, the optimization and solution operations in S3 are repeated based on the current updated traffic status, the charging station queue, and the vehicle's remaining battery power. If there is a feasible charging station with a lower overall cost that meets all constraints, the charging reservation is updated and the vehicle is navigated to the newly selected charging station.

[0016] S5. When the electric vehicle arrives at the charging station finally selected after dynamic optimization, the charging service will be executed according to the latest reservation information. After charging is completed, the destination will be set as the navigation target, and the driving route to the destination will be planned based on real-time traffic information until the final destination is reached.

[0017] Compared with the prior art, the present invention has the following advantages:

[0018] 1. Achieve deep integration of multi-source information from "vehicle-road-station-network". Unlike existing technologies that rely solely on static or partial information (such as considering only charging station status or only planning routes), this solution simultaneously acquires real-time traffic network data, charging station operating status, and power grid load information in the S0 stage, and continuously integrates and updates this information in subsequent stages. This multi-dimensional information collaboration mechanism effectively overcomes the problems of inaccurate reservations and resource mismatches caused by information fragmentation in traditional methods.

[0019] 2. Supports dynamic route replanning and reservation updates throughout the entire process. Existing technologies typically select a route and charging station once before departure and then keep them fixed, making it difficult to cope with sudden traffic congestion, surges in charging station queues, or abnormal battery consumption during the journey. This solution introduces a "node-triggered" re-optimization mechanism in S4, where the global state is reassessed and the reservation and navigation routes are dynamically adjusted every time the vehicle reaches a road network node, significantly enhancing the system's robustness and adaptability.

[0020] 3. An intelligent queue-jumping judgment and compensation mechanism is introduced to balance efficiency and fairness. For users with high urgency (such as extremely low battery or time constraints), this solution embeds a queue-jumping condition evaluation module into the S3. When certain thresholds are met, they are allowed priority access, and the increased utility brought by queue-jumping is included in the overall cost calculation. Simultaneously, reasonable compensation is provided to affected users. Compared to existing queuing strategies that lack priority scheduling or are simply first-come, first-served, this mechanism maintains overall charging order while ensuring the needs of critical users.

[0021] 4. Construct a comprehensive cost optimization model with multiple constraints. The optimization model established in S2 not only considers the user's time and energy costs, but also explicitly introduces hard constraints such as traffic capacity, vehicle accessibility, and the safe operation of the power distribution network, ensuring that the selected charging stations and routes are physically feasible and have no impact on the power grid. Compared with traditional methods that only aim at the shortest path or the lowest electricity price, this model is closer to the actual operating environment and avoids outputting theoretically optimal but practically infeasible solutions.

[0022] 5. Improve the efficiency of charging resource utilization and user travel experience. By dynamically matching the optimal charging station, reducing unnecessary waiting and detours, and minimizing reservation failures due to information lag, this solution effectively alleviates the structural contradiction of "crowded queues" at charging stations and "empty and idle" stations during peak hours. At the same time, it shortens the overall travel time and uncertainty for users, significantly improving the user experience.

[0023] In summary, this method integrates multi-dimensional real-time information from "vehicle-road-station-network", supports dynamic route replanning and intelligent queue-jumping judgment, and achieves a real-time charging reservation mechanism for electric vehicles that minimizes the overall travel cost for users while meeting the constraints of power grid safety and traffic capacity.

[0024] Preferably, in S2, the power distribution network safety operation constraints are verified using the DistFlow power flow algorithm, wherein the active power and reactive power from node i to node i+1 satisfy the following recursive relationships:

[0025]

[0026]

[0027] The expression for the node voltage update calculation is:

[0028]

[0029] In the formula, P i(i+1) Q i(i+1) P represents the active power and reactive power transmitted from node i to node i+1, respectively; i Q i These represent the active power and reactive power of node i, respectively; r i(i+1) x i(i+1) V represents the resistance and reactance of the line from node i to node i+1, respectively; i Let be the voltage at node i;

[0030] in,

[0031] In the formula, P i,load The basic load connected to node i; P c (n) represents the charging power of the nth electric vehicle; N represents the number of electric vehicles connected to the power distribution node i and currently charging at the moment.

[0032] Voltage V at node i i Satisfy the preset upper and lower limit constraints:

[0033] V i min ≤V i ≤V i max ;

[0034] In the formula, V i min V i max These are the minimum and maximum voltages allowed at node i, respectively.

[0035] This setup, by introducing the DistFlow power flow algorithm, enables refined modeling and real-time verification of the distribution network's operating status, effectively supporting the dynamic assurance of grid security during the electric vehicle charging reservation process.

[0036] Preferably, in S2, the traffic capacity constraint is:

[0037] v a ≥30% v max ;

[0038] In the formula, v max The ideal maximum speed for road traffic;

[0039] The vehicle accessibility constraint is:

[0040] B t -l os λ k >0;

[0041] In the formula, B t Indicates the current battery level of the electric vehicle; os Indicates the distance from the current location to the charging station; λ k This indicates the energy consumption per unit distance of an electric vehicle;

[0042] The constraints for safe operation of the power distribution network are:

[0043] P t,i <P thr,i ;

[0044] |N s (i,t)|<7%;

[0045] In the formula, P t,i P represents the load of the i-th charging station at time t; thr,i N represents the load threshold of the i-th charging station at time t; s (i,t) represents the voltage offset rate of the i-th node at time t in the distribution network.

[0046] This setup, by constructing a multi-dimensional constraint system that includes traffic capacity, vehicle accessibility, and the safe operation of the power distribution network, enables refined management and control of the entire process of electric vehicle charging reservation, ensuring user travel efficiency while also taking into account traffic flow and power grid security.

[0047] Preferably, in S0, based on the basic principles of graph theory, G = (N, A) is used to describe the topological characteristics of the traffic network; where N is the set of nodes in the network, i.e., the intersections or starting and ending points of roads; A represents the set of all road segments in the traffic network; a road with starting point i and ending point j is represented as a = (i, j) ∈ A;

[0048] The comprehensive cost includes travel time cost, energy consumption cost, charging cost, queuing time, and the utility increment brought about by queue jumping;

[0049] Comprehensive cost L ev The formula for calculation is:

[0050]

[0051] In the formula, the first term on the right side of the second equal sign is the comprehensive road impedance from the starting point o to the charging station s; the second term is the waiting time and charging time cost; the third term is the charging cost and the increment of queue-jumping utility; and the fourth term is the comprehensive road impedance from the charging station s to the destination D.

[0052] Waiting time for electric vehicles; For estimated charging time; C charge The cost of charging electric vehicles; ω is 0 or 1, ω = 1 when U(k) > 0, otherwise 0; U(k) is the increment function of queue jumping utility; α and β represent the unit time cost and unit energy cost of road travel, respectively; and These are the combined road impedances from the starting point to the charging station and from the charging station to the destination, respectively. This represents the set of decision variables for electric vehicles traveling on roads.

[0053] This setup achieves: 1. Structured modeling and efficient path representation of traffic networks. Utilizing fundamental principles of graph theory, road intersections or start / end points are treated as a set of nodes N, and road segments as a set of edges A, thus transforming the real road network into a computable topological structure. Compared to traditional methods relying on empirical rules or simple distance matching, this modeling approach supports accurate representation of complex urban road networks, providing a solid foundation for subsequent dynamic path search and optimization.

[0054] 2. Comprehensive cost function L ev By explicitly incorporating travel time costs, energy consumption costs, charging fees, waiting times, and the utility increase resulting from queue-jumping, this model represents a shift from a single economic indicator to a multi-dimensional user experience. Compared to traditional strategies that only consider charging prices or route length, this model is closer to actual user perception and helps improve the satisfaction and practicality of reservation results.

[0055] 3. Support for quantitative evaluation and incentive mechanism design for queue-jumping behavior. An incremental function U(k) for queue-jumping utility and a control variable ω are introduced. Under specific conditions, the benefits of queue-jumping are included in the total cost calculation (ω = 1 when U(k) > 0). This mechanism allows high-urgency vehicles to have priority access when necessary, while achieving dynamic balance in resource allocation through cost weighting, thus compensating for the lack of flexible priority scheduling capabilities in existing systems.

[0056] 4. Provide a unified objective function framework for dynamic path planning. The path impedances from the starting point to the charging station and from the charging station to the destination are denoted as follows: and This is integrated into the overall cost, bringing the entire journey under unified optimization considerations. This not only avoids the problem of separating path and charging decisions, but also provides a clear objective for subsequent use of the improved Floyd algorithm to find the global optimum, enhancing the algorithm's convergence and practicality.

[0057] Preferably, the formula for calculating the comprehensive road impedance is:

[0058] S ij =α·t ij +β·z ij ;

[0059] In the formula, α and β represent the unit time cost and unit energy consumption cost of road travel, respectively; t ij Indicates the time required to travel through road ij; z ij This represents the energy consumption required to travel through road ij;

[0060]

[0061] In the formula, t o and C ij These represent the free passage time and capacity of road ij, respectively; q ij This indicates the traffic flow on road ij;

[0062] z ij =h ij ·λ k ;

[0063] In the formula, h ij λ is the length of road ij; k This indicates the energy consumption per unit distance of an electric vehicle;

[0064]

[0065] In the formula: V a The speed of the electric vehicle on road a; k represents the first, second and third level roads.

[0066] This setup has two advantages: 1. It can dynamically reflect the impact of traffic congestion on traffic efficiency. ij A modified free-flow time model is adopted, and the traffic flow q is introduced. ij With road capacity C ij The ratio of [the traffic flow rate] to [the traffic flow rate] is used as a congestion factor, causing travel time to increase non-linearly with increasing traffic load. This modeling approach based on traffic flow theory can effectively capture common congestion phenomena in urban road networks, avoid route misjudgments caused by ignoring real-time traffic conditions, and significantly improve the adaptability of navigation results.

[0067] 2. Consider the differences in energy consumption under different road grades and driving conditions. Establish energy consumption λ per unit mileage according to different road grades (Level 1, Level 2, and Level 3). k An empirical model, and the parameters are related to the driving speed V. a This model reflects the differences in energy consumption characteristics of vehicles under different road conditions. Compared with traditional methods that assume constant energy consumption, this model can more realistically reflect the actual power consumption behavior of electric vehicles in complex urban environments, improving the accuracy of energy prediction.

[0068] 3. Supports path optimization decision-making under multi-factor coupling. By unifying time, energy consumption, and traffic conditions into the road impedance expression, and combining it with formula z... ij =h ij ·λ k The energy consumption calculation based on road length and unit energy consumption realizes the transformation from physical attributes to economic costs. This mechanism provides a reliable data foundation for the subsequent construction of comprehensive cost functions, enabling route selection to not only focus on "speed" but also take into account "energy saving" and "smoothness".

[0069] Preferably, the charging cost C for electric vehicles charge The formula for calculation is:

[0070] C charge = p(t)·ΔB;

[0071] In the formula, P(t) is a function of the electric vehicle charging price and charging service price changing with time; ΔB represents the charging amount.

[0072]

[0073] ΔB=0.8B o -B s ;

[0074] B s =B t -l os ·E o ;

[0075] In the formula, B s Indicates the remaining battery power upon arrival at the charging station; B t Indicates the remaining battery power when a charging request is triggered; os Indicates the distance from the current location to the charging station; E o Indicates the energy consumption per unit mileage of electric vehicles; B o This represents the theoretical maximum charge of an electric vehicle;

[0076] Expected charging time The formula for calculation is:

[0077]

[0078] In the formula, η s For charging efficiency; P s This refers to the power output of the charging station.

[0079] This setup, by constructing a charging cost and time prediction model that integrates dynamic electricity prices, actual electricity demand, and charging efficiency, enables refined modeling of electric vehicle charging behavior, improving users' economic benefits and travel efficiency while ensuring the accuracy of charging services.

[0080] Preferably, the waiting time for electric vehicles for:

[0081]

[0082] In the formula, This is the charging start time; This is the estimated arrival time;

[0083]

[0084] In the formula, t a Indicates the time when an electric vehicle triggers its charging demand; x ij For 0-1 variables, when x ij =1, indicating that the electric vehicle passes through road ij; when x ij =0, indicating that the electric vehicle will not cross the road ij; B s Indicates the remaining battery power upon arrival at the charging station; B t Indicates the remaining battery power when the charging demand is triggered; ΔB represents the charging capacity; l os Indicates the distance from the current location to the charging station; η s =0.9 represents the charging efficiency; P s This refers to the power output of the charging station.

[0085] Charging start time The formula for calculation is:

[0086]

[0087] P c (t arr )=p∈P s ,

[0088]

[0089] In the formula, t o For query time; t arr This is the estimated arrival time; Indicates the fastest idle time of the charging station; C s Let S be the set of charging piles for charging stations s∈S; Indicates the idle time of charging station c; P c (t arr ) indicates that the device is assigned to charging station c and arrives before time t. arr Charging plans; This indicates the query time when the charging station is not occupied; p represents the charging plan for vehicles that have been reserved; P s This indicates the pre-booked charging schedule for charging station s; This indicates the charging duration in the charging plan p; The estimated arrival time in charging plan p; The time that charging stations take up for vehicle departure.

[0090] This setup supports: 1. Spatiotemporal coordinated scheduling of charging plans and detection of resource conflicts. Charging start time. The calculation formula incorporates a complete charging plan matching logic: it determines whether a vehicle has already been reserved. Find the earliest available charging station The charging start time is calculated by combining the charging time. This mechanism enables the allocation of charging resources over time, effectively avoiding resource conflicts caused by multiple vehicles making reservations simultaneously and enhancing the rationality of system scheduling.

[0091] 2. Integrating the time coupling relationship between route planning and charging services. In the calculation formula, through the path variable x ij The energy consumption model will predict the arrival time. By linking the driving route and vehicle status, waiting time calculation is no longer isolated from route selection. This "road-station" linkage modeling method ensures that charging decisions fully consider the time continuity of the entire travel process, improving the overall trip coordination and user experience.

[0092] Preferably, in S2, when determining whether the queue-jumping condition is met, if the urgency of user x's charging is... If the following condition is met, then user x has the right to jump the queue:

[0093]

[0094] In the formula, z represents the number of users ahead of user x in the queue; R risk (t) represents the urgency of charging for users in the queue;

[0095]

[0096] In the formula, SOC cri Indicates the remaining battery power safety threshold; SOC user (t) represents the current battery level of the electric vehicle.

[0097] This setup achieves two key advantages: 1. Establishing a reasonable threshold for queue-jumping decisions, balancing efficiency and order. The queue-jumping condition is set as follows: when user x's urgency level is greater than or equal to two-thirds of the average urgency level of the z users ahead in the queue, they are granted the right to jump the queue. This design avoids system imbalance caused by "absolute priority," allowing high-risk users to access the queue first to prevent them from being stranded, while maintaining a certain degree of queuing fairness, effectively balancing the conflict between efficiency and impartiality.

[0098] 2. Achieve objective quantification and dynamic assessment of charging urgency. This involves quantifying and dynamically evaluating the user's charging urgency level, R. risk (t) is defined as the relative difference between the current remaining battery power and the safety threshold, which can intuitively reflect the risk level of the vehicle approaching battery depletion. Compared with subjective judgment or methods that are simply ranked based on remaining battery power, this model is more physically meaningful and comparable, and supports fair comparison among multiple users.

[0099] 3. Provides a decision-making basis for the incremental utility of queue jumping and the compensation mechanism. The mathematical expression of the queue jumping conditions provides a clear criterion for introducing the "incremental utility of queue jumping" into the subsequent comprehensive cost function. When the conditions are met, the system can include the time savings or safety improvements brought by queue jumping into the total revenue, and at the same time provide corresponding compensation to the affected users (such as delayed service time, points rewards, etc.), thereby constructing a closed-loop priority scheduling mechanism.

[0100] Preferably, in S2, when providing corresponding compensation to users affected by queue jumping, the expected compensation amount Z for user x is calculated based on user x's expected queue jumping position (kth position). com :

[0101]

[0102] z-1≤k≤z;

[0103] In the formula, f i This represents the compensation amount for the i-th user who was cut in line;

[0104]

[0105] In the formula, Δt ps This indicates the original waiting time for electric vehicles. Waiting time after being cut in line The interval; α(t) represents the dynamic queueing compensation coefficient;

[0106] The formula for calculating the dynamic queue-jumping compensation coefficient α(t) is:

[0107] α(t)=α base ·[λ1·L(t)+λ2·(1-R risk(t))]·ε his (t);

[0108] In the formula, λ1 and λ2 are weighting coefficients; α base The basic compensation coefficient; L(t) represents the charging station load index; ε his (t) represents the user's historical queue-jumping format;

[0109]

[0110] In the formula, d represents the number of electric vehicles currently charging; p i (t) represents the charging power of the i-th electric vehicle at time t; P max Indicates the maximum capacity of the charging station; This indicates the number of times someone cuts in line in the past 30 days;

[0111] The formula for calculating the increment of the queue-jumping utility U(k) is:

[0112]

[0113] In the formula, Let x be the initial queuing time. The queuing time for user x after cutting in line.

[0114] This setup achieves two key benefits: 1. It enables a quantitative assessment of the incremental utility of queue-jumping. The incremental utility U(k) of queue-jumping is defined as the weighted sum of the differences in waiting time before and after queue-jumping, minus the cumulative losses suffered by affected users. This model not only measures the time gain of queue-jumping users but also explicitly considers the queuing delays caused to other users, ensuring that the overall cost function truly reflects the net social benefits of queue-jumping.

[0115] 2. Supports dynamic compensation strategies based on system operating status. A dynamic queue-jumping compensation coefficient α(t) is defined, the value of which is determined by the charging station load index L(t) and the user's historical queue-jumping frequency ε. his The value of L(t) and the basic compensation coefficient are jointly determined. When the charging station load is high (L(t) is large), the system tends to increase the compensation to encourage orderly access; while users who frequently cut in line will be penalized due to ε. his (t) increases and is suppressed, avoiding abuse of priority and enhancing the incentive compatibility of the rules.

[0116] 3. Achieve accurate compensation calculation for affected users. Calculate the compensation amount Z for each user who was cut in line. com And the compensation amount for a single user f i , where f iThe compensation is proportional to the amount of waiting time extension and is adjusted through a dynamic coefficient. This compensation mechanism, based on the actual impact, avoids unreasonable allocation through "averaging" or "fixed amounts," thus improving the fairness of compensation and user acceptance.

[0117] Preferably, in step S5, the overall user satisfaction is also calculated:

[0118]

[0119] In the formula, f os For overall satisfaction; The satisfaction levels were for detour distance, queuing time, and charging cost, respectively. ε and ρ are the weighting coefficients of the three satisfaction indicators, respectively. Attached Figure Description

[0120] To make the objectives, technical solutions, and advantages of the invention clearer, the invention will now be described in further detail with reference to the accompanying drawings, wherein:

[0121] Figure 1 Here is a flowchart of the method;

[0122] Figure 2 This is a schematic diagram of the electric vehicle charging reservation service platform architecture in the embodiment.

[0123] Figure 3 This is a schematic diagram of the traffic network topology in the embodiment. Detailed Implementation

[0124] The following detailed explanation illustrates the specific implementation methods:

[0125] Example:

[0126] like Figure 1 , Figure 2 As shown, this embodiment discloses a method for real-time charging reservation of electric vehicles, including the following steps:

[0127] S0. Obtain real-time traffic network data, the operating status of each charging station, and power grid load information. In specific implementation, the above data (such as...) can be obtained through collaboration between a centralized charging station data center (CSDC) and a cloud platform. Figure 2 As shown in the image, it will not be elaborated upon here.

[0128] The traffic network data includes road segment speeds and traffic flow, used to construct the traffic network topology; the operating status of the charging station includes charging pile occupancy and queue information.

[0129] In practical implementation, based on the fundamental principles of graph theory, G = (N, A) is used to describe the topological characteristics of the traffic network; where N is the set of nodes in the network, i.e., the intersections or start and end points of roads; A represents the set of all road segments in the traffic network; a road with start point i and end point j is represented as a = (i, j) ∈ A. The topological structure of the traffic network is as follows: Figure 3 As shown, all roads are two-way streets. To accurately describe the real-time traffic network conditions, the travel time and energy cost of each road are considered as the road's overall impedance.

[0130] This approach enables structured modeling and efficient path representation of traffic networks. Utilizing fundamental principles of graph theory, it treats road intersections or start / end points as the set of nodes N and road segments as the set of edges A, thus transforming the real-world road network into a computable topological structure. Compared to traditional methods relying on empirical rules or simple distance matching, this modeling approach supports accurate representation of complex urban road networks, providing a solid foundation for subsequent dynamic path search and optimization.

[0131] S1. Obtain the initial state information of the electric vehicle that initiated the charging request; wherein, the initial state information includes the starting point, the destination, the departure time, and the remaining battery power.

[0132] In practice, when the electric vehicle's battery level falls below a safe threshold, a charging demand is triggered, as shown in the following formula:

[0133] B t ≤0.2B o Or B t ≤l oD ·λ k ;

[0134] In the formula, B t Indicates the current battery level of the electric vehicle; B o Indicates the theoretical electric capacity of an electric vehicle; l oD This indicates the distance from the point where the charging demand is triggered to the destination.

[0135] S2. Based on the information obtained from S0 and S1, an optimization model is constructed with the goal of minimizing the overall cost. The optimization model introduces traffic capacity constraints, vehicle accessibility constraints, and power distribution network safety operation constraints to limit the feasible selection of charging stations and route combinations.

[0136] In practical implementation, the power distribution network safety operation constraints are verified using the DistFlow power flow algorithm, where the active and reactive power from node i to node i+1 satisfy the following recursive relationships:

[0137]

[0138] The expression for the node voltage update calculation is:

[0139]

[0140] In the formula, P i(i+1) Q i(i+1) P represents the active power and reactive power transmitted from node i to node i+1, respectively; i Q i These represent the active power and reactive power of node i, respectively; r i(i+1) x i(i+1) V represents the resistance and reactance of the line from node i to node i+1, respectively; i Let be the voltage at node i;

[0141] in,

[0142] In the formula, P i,load The basic load connected to node i; P c (n) represents the charging power of the nth electric vehicle; N represents the number of electric vehicles connected to the power distribution node i and currently charging at the moment.

[0143] Voltage V at node i i Satisfy the preset upper and lower limit constraints:

[0144] V i min ≤V i ≤V i max ;

[0145] In the formula, V i min V i max These are the minimum and maximum voltages allowed at node i, respectively.

[0146] The traffic capacity constraint is as follows:

[0147] v a ≥30% v max ;

[0148] In the formula, v max The ideal maximum speed for road traffic;

[0149] The vehicle accessibility constraint is:

[0150] B t -l os λ k >0;

[0151] In the formula, B t Indicates the current battery level of the electric vehicle; os Indicates the distance from the current location to the charging station; λ kThis indicates the energy consumption per unit distance of an electric vehicle;

[0152] The constraints for safe operation of the power distribution network are:

[0153] P t,i <P thr,i ;

[0154] |N s (i,t)|<7%;

[0155] In the formula, P t,i P represents the load of the i-th charging station at time t; thr,i N represents the load threshold of the i-th charging station at time t; s (i,t) represents the voltage offset rate of the i-th node at time t in the distribution network.

[0156] The comprehensive cost includes travel time cost, energy consumption cost, charging cost, queuing time, and the utility increment brought about by queue jumping;

[0157] In practical implementation, the comprehensive cost L ev The formula for calculation is:

[0158]

[0159] In the formula, the first term on the right side of the second equal sign is the comprehensive road impedance from the starting point o to the charging station s; the second term is the waiting time and charging time cost; the third term is the charging cost and the increment of queue-jumping utility; and the fourth term is the comprehensive road impedance from the charging station s to the destination D.

[0160] Waiting time for electric vehicles; For estimated charging time; C charge The cost of charging electric vehicles; ω is 0 or 1, ω = 1 when U(k) > 0, otherwise 0; U(k) is the increment function of queue jumping utility; α and β represent the unit time cost and unit energy cost of road travel, respectively; and These are the combined road impedances from the starting point to the charging station and from the charging station to the destination, respectively. This represents the set of decision variables for electric vehicles traveling on roads.

[0161] Comprehensive cost function L evBy explicitly incorporating travel time costs, energy consumption costs, charging fees, queuing time, and the utility increment resulting from queue-jumping, this model shifts from a single economic indicator to a multi-dimensional user experience. Compared to traditional strategies that only consider charging prices or path length, this model is closer to actual user perception, helping to improve the satisfaction and practicality of reservation results. Furthermore, by introducing a queue-jumping utility increment function U(k) and a control variable ω, the benefits of queue-jumping are included in the total cost calculation under specific conditions (ω = 1 when U(k) > 0). This mechanism allows high-urgency vehicles to have priority access when necessary, while achieving dynamic balance in resource allocation through cost weighting, compensating for the lack of flexible priority scheduling capabilities in existing systems. In addition, the path impedances from the origin to the charging station and from the charging station to the destination are denoted as... and This is integrated into the overall cost, bringing the entire journey under unified optimization considerations. This not only avoids the problem of separating path and charging decisions, but also provides a clear objective for subsequent use of the improved Floyd algorithm to find the global optimum, enhancing the algorithm's convergence and practicality.

[0162] In practical implementation, the formula for calculating the comprehensive road impedance is:

[0163] S ij =α·t ij +β·z ij ;

[0164] In the formula, α and β represent the unit time cost and unit energy consumption cost of road travel, respectively; t ij Indicates the time required to travel through road ij; z ij This represents the energy consumption required to travel through road ij;

[0165]

[0166] In the formula, t o and C ij These represent the free passage time and capacity of road ij, respectively; q ij This indicates the traffic flow on road ij;

[0167] z ij =h ij ·λ k ;

[0168] In the formula, h ij λ is the length of road ij; k This indicates the energy consumption per unit distance of an electric vehicle;

[0169]

[0170] In the formula: V aThe speed of the electric vehicle on road a; k represents the first, second and third level roads.

[0171] t ij A modified free-flow time model is adopted, and the traffic flow q is introduced. ij With road capacity C ij The ratio of [the ratio of traffic flow to congestion] is used as a congestion factor, causing travel time to increase non-linearly with increasing traffic load. This modeling method based on traffic flow theory can effectively capture common congestion phenomena in urban road networks, avoid route misjudgments caused by ignoring real-time traffic conditions, and significantly improve the adaptability of navigation results. In addition, energy consumption per unit mileage λ is established according to different road levels (Level 1, Level 2, and Level 3). k An empirical model, and the parameters are related to the driving speed V. a This correlation reflects the differences in vehicle energy consumption characteristics under various road conditions. Compared to traditional methods that assume constant energy consumption, this model more realistically reflects the actual energy consumption behavior of electric vehicles in complex urban environments, improving energy prediction accuracy. Furthermore, by unifying time, energy consumption, and traffic conditions into the road impedance expression, and combining it with formula z... ij =h ij ·λ k The energy consumption calculation based on road length and unit energy consumption realizes the transformation from physical attributes to economic costs. This mechanism provides a reliable data foundation for the subsequent construction of comprehensive cost functions, enabling route selection to not only focus on "speed" but also take into account "energy saving" and "smoothness".

[0172] Among them, the cost of electric vehicle charging C charge The formula for calculation is:

[0173] C charge = p(t)·ΔB;

[0174] In the formula, P(t) is a function of the electric vehicle charging price and charging service price changing with time; ΔB represents the charging amount.

[0175]

[0176] ΔB=0.8B o -B s ;

[0177] B s =B t -l os ·E o ;

[0178] In the formula, B s Indicates the remaining battery power upon arrival at the charging station; B t Indicates the remaining battery power when a charging request is triggered; osIndicates the distance from the current location to the charging station; E o Indicates the energy consumption per unit mileage of electric vehicles; B o This represents the theoretical maximum charge of an electric vehicle;

[0179] Expected charging time The formula for calculation is:

[0180]

[0181] In the formula, η s For charging efficiency; P s This refers to the power output of the charging station.

[0182] Electric vehicle waiting time for:

[0183]

[0184] In the formula, This is the charging start time; This is the estimated arrival time;

[0185]

[0186] In the formula, t a Indicates the time when an electric vehicle triggers its charging demand; x ij For 0-1 variables, when x ij =1, indicating that the electric vehicle passes through road ij; when x ij =0, indicating that the electric vehicle will not cross the road ij; B s Indicates the remaining battery power upon arrival at the charging station; B t Indicates the remaining battery power when the charging demand is triggered; ΔB represents the charging capacity; l os Indicates the distance from the current location to the charging station; η s =0.9 represents the charging efficiency; P s This refers to the power output of the charging station.

[0187] Charging start time The formula for calculation is:

[0188]

[0189]

[0190] P c (t arr )=p∈P s ,

[0191]

[0192] In the formula, to For query time; t arr This is the estimated arrival time; Indicates the fastest idle time of the charging station; C s Let S be the set of charging piles for charging stations s∈S; Indicates the idle time of charging station c; P c (t arr ) indicates that the device is assigned to charging station c and arrives before time t. arr Charging plans; This indicates the query time when the charging station is not occupied; p represents the charging plan for vehicles that have been reserved; P s This indicates the pre-booked charging schedule for charging station s; This indicates the charging duration in the charging plan p; The estimated arrival time in charging plan p; The time that charging stations take up for vehicle departure.

[0193] Charging start time The calculation formula incorporates a complete charging plan matching logic: it determines whether a vehicle has already been reserved. Find the earliest available charging station The charging start time is calculated by combining the charging time. This mechanism enables the allocation of charging resources over time, effectively avoiding resource conflicts caused by multiple vehicles making simultaneous reservations and enhancing the rationality of system scheduling. In addition, In the calculation formula, through the path variable x ij The energy consumption model will predict the arrival time. By linking the driving route and vehicle status, waiting time calculation is no longer isolated from route selection. This "road-station" linkage modeling method ensures that charging decisions fully consider the time continuity of the entire travel process, improving the overall trip coordination and user experience.

[0194] To better illustrate the derivation process of the comprehensive cost in this method, the following explanation is provided:

[0195] When an electric vehicle needs charging, it selects a suitable charging station through the CSDC and cloud platform, makes a reservation, and changes its destination to the reserved charging station. Upon arrival at the charging station, it queues for charging according to the reservation queue. After charging, it proceeds to its initial destination. The total charging cost consists of three parts: the time and energy cost required for the electric vehicle to reach the charging station (L1); the sum of the waiting time, charging time, and incremental utility from charging and queuing at the charging station (L2); and finally, the time and energy cost required to return to the initial destination after charging (L3). L1 and L3 are derived from equation S. ij =α·t ij +β·zij Calculated.

[0196] When the electric vehicle arrives at charging station c, the charging cost C can be calculated. charge As shown in the following formula.

[0197] C charge = p(t)·ΔB;

[0198] The charging cost is mainly determined by the charging price and the amount of electricity charged. The charging price includes the electricity price and the charging service fee. This invention takes the Beijing charging electricity price as an example, as shown in the following formula. P(t) is a function of the electric vehicle charging price and the charging service price changing over time.

[0199]

[0200] Therefore, the charging cost L2 of an electric vehicle at a charging station can be calculated as follows:

[0201]

[0202] In summary, this method establishes an objective function considering time cost, energy cost, queue-jumping utility increment, and various charging costs, and aims to minimize the total cost for charging navigation. Therefore, the total road travel cost of the electric vehicle is:

[0203]

[0204] In practice, when determining whether the conditions for queue jumping are met, if user x's charging urgency level is... If the following condition is met, then user x has the right to jump the queue:

[0205]

[0206] In the formula, z represents the number of users ahead of user x in the queue; R risk (t) represents the urgency of charging for users in the queue;

[0207]

[0208] In the formula, SOC cri Indicates the remaining battery power safety threshold; SOC user (t) represents the current battery level of the electric vehicle.

[0209] Thus, the queue-jumping condition is set as follows: when user x's urgency level is greater than or equal to two-thirds of the average urgency level of the z users ahead in the queue, they are granted the right to jump the queue. This design avoids system imbalance caused by "absolute priority," allowing high-risk users to access the queue first to prevent breakdowns while maintaining a certain degree of queuing fairness, effectively balancing the contradiction between efficiency and impartiality. In addition, the user's charging urgency level R...risk (t) is defined as the relative difference between the current remaining battery power and the safety threshold, which can intuitively reflect the risk level of the vehicle approaching battery depletion. Compared with subjective judgment or methods that are simply ranked based on remaining battery power, this model is more physically meaningful and comparable, and supports fair comparison among multiple users.

[0210] It should be noted that in practical applications, for emergency charging needs of electric public service vehicles performing urgent tasks, such as ambulances, fire trucks, and roadside assistance vehicles, a special priority queue-jumping mechanism can be established: charging stations grant emergency vehicles the highest charging authority and fully record the occupancy time of charging stations, details of compensation fees, and a list of affected users, forming a quarterly special report submitted to the regulatory authorities for review. The government provides public financial compensation through tax adjustments or electricity price subsidies, forming a social cost-sharing mechanism.

[0211] When compensating users affected by queue jumping, the expected compensation amount Z for user x is calculated based on user x's expected queue jump position (kth position). com :

[0212]

[0213] z-1≤k≤z;

[0214] In the formula, f i This represents the compensation amount for the i-th user who was cut in line;

[0215]

[0216] In the formula, Δt ps This indicates the original waiting time for electric vehicles. Waiting time after being cut in line The interval; α(t) represents the dynamic queueing compensation coefficient;

[0217] The formula for calculating the dynamic queue-jumping compensation coefficient α(t) is:

[0218] α(t)=α base ·[λ1·L(t)+λ2·(1-R risk (t))]·ε his (t);

[0219] In the formula, λ1 and λ2 are weighting coefficients; α base The basic compensation coefficient has different values ​​during peak, valley, and normal load periods, as shown in Table 1; L(t) represents the charging station load index; ε his (t) represents the user's historical queue-jumping format.

[0220] Table 1 Basic Compensation Coefficients for Different Time Periods

[0221]

[0222]

[0223] In the formula, d represents the number of electric vehicles currently charging; p i (t) represents the charging power of the i-th electric vehicle at time t; P max Indicates the maximum capacity of the charging station; This indicates the number of times someone cuts in line in the past 30 days;

[0224] The formula for calculating the increment of the queue-jumping utility U(k) is:

[0225]

[0226] In the formula, Let x be the initial queuing time. The queuing time for user x after cutting in line.

[0227] Based on the bounded rationality assumption in economics, user behavior decisions in charging scenarios follow a cost-benefit equilibrium principle. That is, a user will only trigger a queue-jumping decision when the estimated increase in time value gained from the queue-jumping action exceeds the resulting economic cost. Therefore, the maximum value of the user's utility increment U(k) is calculated. If U(k) > 0, user x chooses to jump the queue; otherwise, they do not. Simultaneously, CSDC updates the reservation queue based on the user's queue-jumping decision and updates the user's credit score.

[0228] Thus, the utility increment U(k) for queue jumping is defined as the weighted sum of the waiting time differences before and after queue jumping, minus the cumulative losses of affected users. This model not only measures the time gain of queue-jumping users but also explicitly considers the queuing delays caused to other users, ensuring that the comprehensive cost function truly reflects the net social benefits of queue-jumping behavior. Furthermore, a dynamic queue-jumping compensation coefficient α(t) is defined, its value determined by the charging station load index L(t) and the user's historical queue-jumping frequency ε. his The value of L(t) and the basic compensation coefficient are jointly determined. When the charging station load is high (L(t) is large), the system tends to increase the compensation to encourage orderly access; while users who frequently cut in line will be penalized due to ε. his (t) increases but is suppressed, preventing abuse of priority and enhancing the incentive compatibility of the rules. Furthermore, the compensation amount Z for users who are cut in line is calculated separately. com And the compensation amount for a single user f i , where f i The compensation is proportional to the amount of waiting time extension and is adjusted through a dynamic coefficient. This compensation mechanism, based on the actual impact, avoids unreasonable allocation through "averaging" or "fixed amounts," thus improving the fairness of compensation and user acceptance.

[0229] S3. Using the improved dynamic Floyd algorithm, based on the optimization model built in S2, calculate the corresponding comprehensive cost for each feasible charging station, and solve for the charging station with the lowest comprehensive cost and its corresponding navigation path; and make corresponding charging reservations.

[0230] In calculating the overall cost of each candidate charging station, it is simultaneously assessed whether the electric vehicle that initiated the charging request meets the conditions for queuing at that station. If it does, the queuing utility increment is included in the overall cost calculation, and corresponding compensation is provided to users affected by queuing.

[0231] S4. During the journey of the electric vehicle to the reserved charging station, whenever it reaches any node in the traffic network, the optimization and solution operations in S3 are repeated based on the current updated traffic status, the charging station queue, and the vehicle's remaining battery power. If there is a feasible charging station with a lower overall cost that meets all constraints, the charging reservation is updated and the vehicle is navigated to the newly selected charging station.

[0232] S5. When the electric vehicle arrives at the charging station finally selected after dynamic optimization, the charging service will be executed according to the latest reservation information. After charging is completed, the destination will be set as the navigation target, and the driving route to the destination will be planned based on real-time traffic information until the final destination is reached.

[0233] In practice, overall user satisfaction is also calculated:

[0234]

[0235] In the formula, f os For overall satisfaction; The satisfaction levels were for detour distance, queuing time, and charging cost, respectively. ε and ρ are the weighting coefficients of the three satisfaction indicators, respectively. In practice, ω = 0.2, ε = 0.5, and ρ = 0.3.

[0236] In practice, S5 continuously updates the charging pile occupancy status, user credit score, and queue-jumping compensation records after the charging service is completed, forming a closed-loop feedback mechanism to optimize the subsequent electric vehicle reservation and scheduling strategy.

[0237] Unlike existing technologies that rely solely on static or localized information (such as considering only charging station status or only planning routes), this solution simultaneously acquires real-time traffic network data, charging station operating status, and grid load information in stage S0, and continuously integrates and updates this information in subsequent stages. This multi-dimensional information collaboration mechanism effectively overcomes the problems of inaccurate reservations and resource mismatches caused by information fragmentation in traditional methods. Furthermore, existing technologies typically fix the route and charging station selection once before departure, making it difficult to cope with sudden traffic congestion, surges in charging station queues, or abnormal power consumption during the journey. This solution introduces a "node-triggered" re-optimization mechanism in stage S4, where the global state is reassessed and the reservation and navigation routes are dynamically adjusted each time the vehicle reaches a road network node, significantly enhancing the system's robustness and adaptability.

[0238] For users with high urgency (such as extremely low battery or time constraints), this solution embeds a queue-jumping condition evaluation module in S3. When a specific threshold is met, they are allowed priority access, and the utility increment brought by queue-jumping is included in the overall cost calculation. Simultaneously, reasonable compensation is provided to affected users. Compared to existing queuing strategies that lack priority scheduling or are simply first-come, first-served, this mechanism maintains overall charging order while ensuring the needs of critical users. Furthermore, the optimization model established in S2 not only considers users' time and energy costs but also explicitly introduces hard constraints such as traffic capacity, vehicle accessibility, and the safe operation of the power distribution network, ensuring that the selected charging stations and routes are physically feasible and have no impact on the power grid. Compared to traditional methods that only aim for the shortest path or the lowest electricity price, this model is closer to the actual operating environment, avoiding outputting theoretically optimal but practically infeasible solutions. Furthermore, by dynamically matching the optimal charging station, reducing unnecessary waiting and detours, and minimizing reservation failures due to information lag, this solution effectively alleviates the structural contradiction of "crowded queues" at charging stations and "empty and idle" stations during peak hours, while shortening the overall travel time and uncertainty for users and significantly improving the user experience.

[0239] This method integrates multi-dimensional real-time information from "vehicle-road-station-network", supports dynamic route replanning and intelligent queue-jumping judgment, and achieves a real-time electric vehicle charging reservation mechanism that minimizes the overall travel cost for users while meeting the constraints of power grid safety and traffic capacity.

[0240] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit the technical solutions. Those skilled in the art should understand that any modifications or equivalent substitutions to the technical solutions of the present invention without departing from the spirit and scope of the present invention should be covered within the scope of the claims of the present invention.

Claims

1. A method for real-time charging reservation for electric vehicles, characterized in that, Includes the following steps: S0. Obtain real-time traffic network data, the operating status of each charging station, and power grid load information; wherein, the traffic network data includes road segment speed and traffic flow, used to construct the traffic network topology; the operating status of the charging stations includes charging pile occupancy and queue information; S1. Obtain the initial state information of the electric vehicle that initiated the charging request; wherein, the initial state information includes the starting point, the destination, the departure time, and the remaining battery power; S2. Based on the information obtained from S0 and S1, an optimization model is constructed with the goal of minimizing the overall cost. The optimization model introduces traffic capacity constraints, vehicle accessibility constraints, and power distribution network safety operation constraints to limit the feasible selection of charging stations and route combinations. S3. Using the improved dynamic Floyd algorithm, based on the optimization model built in S2, calculate the corresponding comprehensive cost for each feasible charging station, and solve for the charging station with the lowest comprehensive cost and its corresponding navigation path; and make corresponding charging reservations. In calculating the overall cost of each candidate charging station, it is simultaneously assessed whether the electric vehicle that initiated the charging request meets the conditions for queuing at that station. If it does, the queuing utility increment is included in the overall cost calculation, and corresponding compensation is provided to users affected by queuing. S4. During the journey of the electric vehicle to the reserved charging station, whenever it reaches any node in the traffic network, the optimization and solution operations in S3 are repeated based on the current updated traffic status, the charging station queue, and the vehicle's remaining battery power. If there is a feasible charging station with a lower overall cost that meets all constraints, the charging reservation is updated and the vehicle is navigated to the newly selected charging station. S5. When the electric vehicle arrives at the charging station finally selected after dynamic optimization, the charging service will be executed according to the latest reservation information. After charging is completed, the destination will be set as the navigation target, and the driving route to the destination will be planned based on real-time traffic information until the final destination is reached.

2. The method for real-time charging reservation of electric vehicles as described in claim 1, characterized in that: In S2, the power distribution network safety operation constraints are verified using the DistFlow power flow algorithm, where the active and reactive power from node i to node i+1 satisfy the following recursive relationships: The expression for the node voltage update calculation is: In the formula, P i(i+1) Q i(i+1) P represents the active power and reactive power transmitted from node i to node i+1, respectively; i Q i These represent the active power and reactive power of node i, respectively; r i(i+1) x i(i+1) V represents the resistance and reactance of the line from node i to node i+1, respectively; i Let be the voltage at node i; in, In the formula, P i,load The basic load connected to node i; P c (n) represents the charging power of the nth electric vehicle; N represents the number of electric vehicles connected to the power distribution node i and currently charging at the moment. Voltage V at node i i Satisfy the preset upper and lower limit constraints: In i min ≤V i ≤V i max ; In the formula, V i min V i max These are the minimum and maximum voltages allowed at node i, respectively.

3. The method for real-time charging reservation of electric vehicles as described in claim 2, characterized in that: In S2, the traffic capacity constraint is: v a ≥30%v max ; In the formula, v max The ideal maximum speed for road traffic; The vehicle accessibility constraint is: B t -l os l k >0; In the formula, B t Indicates the current battery level of the electric vehicle; os Indicates the distance from the current location to the charging station; λ k This indicates the energy consumption per unit distance of an electric vehicle; The constraints for safe operation of the power distribution network are: P t,i <P thr,i ; |N s (i,t)|<7%; In the formula, P t,i P represents the load of the i-th charging station at time t; thr,i N represents the load threshold of the i-th charging station at time t; s (i,t) represents the voltage offset rate of the i-th node at time t in the distribution network.

4. The method for real-time charging reservation of electric vehicles as described in claim 3, characterized in that: In S0, based on the basic principles of graph theory, G = (N, A) is used to describe the topological characteristics of the transportation network. Where N is the set of nodes in the road network, that is, the intersections or start and end points of roads; A represents the set of all road segments in the traffic network; a road with a starting point i and an ending point j is represented as a = (i,j)∈A; The comprehensive cost includes travel time cost, energy consumption cost, charging cost, queuing time, and the utility increment brought about by queue jumping; Comprehensive cost L ev The formula for calculation is: In the formula, the first term on the right side of the second equal sign is the comprehensive road impedance from the starting point o to the charging station s; the second term is the waiting time and charging time cost; the third term is the charging cost and the increment of queue-jumping utility; and the fourth term is the comprehensive road impedance from the charging station s to the destination D. Waiting time for electric vehicles; For estimated charging time; C charge The cost of charging electric vehicles; ω is 0 or 1, ω = 1 when U(k) > 0, otherwise 0; U(k) is the increment function of queue jumping utility; α and β represent the unit time cost and unit energy cost of road travel, respectively; and These are the combined road impedances from the starting point to the charging station and from the charging station to the destination, respectively. This represents the set of decision variables for electric vehicles traveling on roads.

5. The method for real-time charging reservation of electric vehicles as described in claim 4, characterized in that: The formula for calculating the comprehensive road impedance is: S ij =α·t ij +β·z ij ; In the formula, α and β represent the unit time cost and unit energy consumption cost of road travel, respectively; t ij Indicates the time required to travel through road ij; z ij This represents the energy consumption required to travel through road ij; In the formula, t o and C ij These represent the free passage time and capacity of road ij, respectively; q ij This indicates the traffic flow on road ij; z ij =h ij ·λ k ; In the formula, h ij λ is the length of road ij; k This indicates the energy consumption per unit distance of an electric vehicle; In the formula: V a The speed of the electric vehicle on road a; k represents the first, second and third level roads.

6. The method for real-time charging reservation of electric vehicles as described in claim 5, characterized in that: Electric vehicle charging cost C charge The formula for calculation is: C charge =p(t)·ΔB; In the formula, P(t) is a function of the electric vehicle charging price and charging service price changing with time; ΔB represents the charging amount. ΔB=0.8B o -B s ; B s =B t -l os ·E o ; In the formula, B s Indicates the remaining battery power upon arrival at the charging station; B t Indicates the remaining battery power when a charging request is triggered; os Indicates the distance from the current location to the charging station; E o Indicates the energy consumption per unit mileage of electric vehicles; B o This represents the theoretical maximum charge of an electric vehicle; Expected charging time The formula for calculation is: In the formula, η s For charging efficiency; P s This refers to the power output of the charging station.

7. The method for real-time charging reservation of electric vehicles as described in claim 6, characterized in that: Electric vehicle waiting time for: In the formula, This is the charging start time; This is the estimated arrival time; In the formula, t a Indicates the time when an electric vehicle triggers its charging demand; x ij For 0-1 variables, when x ij =1, indicating that the electric vehicle passes through road ij; when x ij =0, indicating that the electric vehicle will not cross the road ij; B s Indicates the remaining battery power upon arrival at the charging station; B t Indicates the remaining battery power when the charging demand is triggered; ΔB represents the charging capacity; l os Indicates the distance from the current location to the charging station; η s =0.9 represents the charging efficiency; P s The power of the charging pile; Charging start time The formula for calculation is: P c (t arr )=p∈P s , In the formula, t o For query time; t arr This is the estimated arrival time; Indicates the fastest idle time of the charging station; C s Let S be the set of charging piles for charging stations s∈S; Indicates the idle time of charging station c; P c (t arr ) indicates that the device is assigned to charging station c and arrives before time t. arr Charging plans; This indicates the query time when the charging station is not occupied; p represents the charging plan for vehicles that have been reserved; P s This indicates the pre-booked charging schedule for charging station s; This indicates the charging duration in the charging plan p; The estimated arrival time in charging plan p; The time that charging stations take up for vehicle departure.

8. The method for real-time charging reservation of electric vehicles as described in claim 7, characterized in that: In S2, when determining whether the queue-jumping condition is met, if user x's charging urgency level... If the following condition is met, then user x has the right to jump the queue: In the formula, z represents the number of users ahead of user x in the queue; R risk (t) represents the urgency of charging for users in the queue; In the formula, SOC cri Indicates the remaining battery power safety threshold; SOC user (t) represents the current battery level of the electric vehicle.

9. The method for real-time charging reservation of electric vehicles as described in claim 8, characterized in that: In S2, when compensating users affected by queue jumping, the expected compensation amount Z for user x is calculated based on user x's anticipated queue jump to the kth position. com : z-1≤k≤z; In the formula, f i This represents the compensation amount for the i-th user who was cut in line; where, Δt ps represents the original queuing waiting time of the electric vehicle and the queuing waiting time after being cut in line interval; α(t) represents the dynamic cut-in-line compensation coefficient; The formula for calculating the dynamic queue-jumping compensation coefficient α(t) is: α(t)=α base ·[λ1·L(t)+λ2·(1-R risk (t))]·e his (t); In the formula, λ1 and λ2 are weighting coefficients; α base The basic compensation coefficient; L(t) represents the charging station load index; ε his (t) represents the user's historical queue-jumping format; In the formula, d represents the number of electric vehicles currently charging; p i (t) represents the charging power of the i-th electric vehicle at time t; P max Indicates the maximum capacity of the charging station; This indicates the number of times someone cuts in line in the past 30 days; The formula for calculating the increment of the queue-jumping utility U(k) is: In the formula, Let x be the initial queuing time. The queuing time for user x after cutting in line.

10. The method for real-time charging reservation of electric vehicles as described in claim 9, characterized in that: In S5, overall user satisfaction is also calculated: In the formula, f os For overall satisfaction; The satisfaction levels were for detour distance, queuing time, and charging cost, respectively. ε and ρ are the weighting coefficients of the three satisfaction indicators, respectively.

Images