A method, system, and storage medium for aggregating the regulation capabilities of electric vehicle clusters based on dynamic off-site power boundaries.

By constructing dynamic off-site power boundaries and consistency of state-of-charge demand schedule, the problems of user behavior differences and fixed constraints in the aggregation of electric vehicle cluster regulation capabilities are solved, achieving accurate and executable representation of electric vehicle cluster regulation capabilities and improving the control effect of electric vehicle clusters in virtual power plant scheduling and demand response.

CN122865100BActive Publication Date: 2026-11-17STATE GRID JIANGSU ELECTRIC VEHICLE SERVICE CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202611340811.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2026-09-01
Publication Date
2026-11-17
Estimated Expiration
2046-09-01

AI Technical Summary

Technical Problem

Existing methods for aggregating the regulation capacity of electric vehicle clusters fail to effectively reflect differences in user behavior. Fixed off-site power constraints limit the adjustable space, computational complexity is high, and energy trajectories across time periods are unreliable, leading to distorted regulation capacity estimation and underutilization.

Method used

By constructing a dynamic off-site power boundary, distinguishing between expected off-site power and minimum guaranteed off-site power, and combining the consistency of state-of-charge demand progress, the charging and discharging power and cumulative energy boundaries of the electric vehicle cluster are aggregated to form a unified and executable characterization of regulation capability.

Benefits of technology

It improves the accuracy and feasibility of electric vehicle cluster regulation capabilities, dynamically adjusts off-site power constraints, reduces computational complexity, ensures the feasibility of cross-time period energy trajectories, and enhances the control effect of electric vehicle clusters in virtual power plant scheduling and demand response.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122865100B_ABST
    Figure CN122865100B_ABST
Patent Text Reader

Abstract

This invention relates to a method for aggregating the regulation capacity of electric vehicle clusters based on dynamic off-grid power boundaries. The method includes the following steps: acquiring vehicle entry / exit times, state of charge (SBC), charging / discharging parameters, and off-grid power requirements, and dividing the vehicles into clusters; constructing a dynamic off-grid power boundary for each vehicle based on the expected off-grid power and the minimum guaranteed off-grid power; aggregating the charging / discharging power boundaries of the vehicle clusters based on the consistency of SBC demand progress; and aggregating the upper and lower energy envelopes of each vehicle to construct the cumulative energy boundary of the vehicle clusters, thus forming the aggregation result of the electric vehicle cluster regulation capacity. This invention also discloses a system and related storage media using this method. This invention comprehensively considers user behavior characteristics such as vehicle entry / exit times, exit times, dwell time, initial SBC, and off-grid power requirements, grouping vehicles with similar behavior patterns and regulation needs into the same cluster, reducing intra-cluster differences, and improving the accuracy of subsequent regulation capacity aggregation results.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to a charging method for electric vehicle clusters, specifically a method for aggregating the adjustment capabilities of electric vehicle clusters based on dynamic off-site power boundaries. Background Technology

[0002] With the continuous advancement of new power system construction and the large-scale integration of electric vehicles, electric vehicles, possessing both controllable load and mobile energy storage attributes, can provide peak shaving, demand response, and power regulation services to the power system through orderly charging, delayed charging, and reverse discharge from vehicles to the grid. Virtual power plants or electric vehicle aggregators can organize and schedule a large number of spatially dispersed electric vehicles in a unified manner, transforming the relatively small charging and discharging capacity of individual vehicles into a cluster regulation resource that can be called upon by the grid or the electricity market.

[0003] Before electric vehicle clusters participate in dispatching, it is necessary to aggregate the charging power, discharging power, and cumulative energy adjustment range that the cluster can provide during each dispatching period, based on the access status, current state of charge, battery capacity, charging and discharging power, dwell time, and off-site energy demand of each vehicle. The resulting cluster adjustment capability boundary is an important foundation for formulating charging and discharging plans, conducting demand response, and performing virtual power plant operation control. If the aggregated boundary is larger than the actual executable range of the vehicles, dispatching instructions may not be able to be allocated to specific vehicles; if the aggregated boundary is too conservative, it will result in insufficient utilization of the adjustable capacity of electric vehicles.

[0004] Electric vehicle (EV) users exhibit significant differences and randomness in their charging behavior. Different vehicles show substantial variations in their grid connection time, grid disconnection time, dwell time, initial state of charge (SOC), and off-grid charging demand. Furthermore, different vehicle models possess varying battery capacities, maximum charging / discharging power, charging / discharging efficiencies, and safe SOC ranges. Residential commuter vehicles, office park vehicles, public fast-charging vehicles, and commercial vehicles exhibit different access patterns and charging urgency levels. Treating all vehicles as adjustable resources with identical characteristics weakens the portrayal of user behavior and vehicle physical capabilities, leading to distorted estimates of cluster adjustment capabilities. Existing research also indicates that the adjustable boundary of an EV cluster is influenced by multiple factors, including vehicle physical parameters, grid connection / disconnection time, off-grid charging demand, and user willingness to respond.

[0005] Existing methods for aggregating the regulation capacity of electric vehicles mainly include direct superposition of vehicle rated power, equivalent energy storage modeling, and precise aggregation based on the feasible region of a single vehicle. The direct superposition method of vehicle rated power is computationally simple, but it typically does not fully consider whether the vehicle is online, its remaining dwell time, current battery status, and off-site battery demand. It easily includes currently unavailable charging and discharging power in the cluster regulation capacity calculation, leading to an overestimation of the aggregation result. The equivalent energy storage method describes the cluster's adjustable range through equivalent power and energy boundaries, but it usually uses uniform or fixed off-site battery constraints, making it difficult to reflect the differentiated acceptance range of different users regarding the degree of reduction in off-site battery power when receiving compensation or participating in the response.

[0006] Accurate aggregation methods based on single-vehicle feasible regions can preserve vehicle operational constraints relatively completely, for example, by obtaining the cluster feasible region through polyhedral operations or Minkowski sums. However, as the number of vehicles and scheduling periods increases, the dimensions and number of constraints in the aggregated feasible region grow rapidly, resulting in high computational complexity, which is difficult to meet the needs of rapid modeling and rolling updates of large-scale electric vehicle clusters. On the other hand, if only the charging and discharging power boundary of a single time period is used to describe the cluster's adjustment capability without simultaneously constructing the cumulative energy boundary, the feasibility of cross-time period power trajectory cannot be guaranteed. This may lead to excessive postponement of charging tasks, causing vehicles to fail to meet the minimum off-grid power requirement even when charging at maximum power near the time of grid disconnection.

[0007] Furthermore, existing methods often use the user's desired departure charge level as a fixed constraint, requiring the vehicle to reach a preset desired state of charge upon departure. While this ensures user travel needs, it fails to distinguish between the user's desired departure charge level and the minimum guaranteed departure charge level, and it doesn't fully utilize the undercharge capacity allowed by the user under a given response mechanism. If the user is willing to accept some adjustment in departure charge level, a fixed departure charge level constraint limits the release of delayed charging and reverse discharge capabilities; conversely, directly relaxing the departure charge level constraint may result in the vehicle's departure charge level falling below the user's basic travel needs. Therefore, it is necessary to construct an adjustable dynamic departure charge level boundary between the desired departure charge level and the minimum guaranteed departure charge level.

[0008] In summary, existing electric vehicle cluster regulation capability aggregation technologies have the following main shortcomings:

[0009] 1) Existing methods do not adequately consider the behavioral differences of electric vehicle users, such as the time of entry and exit from the grid, the duration of stay, and the power demand when leaving the grid. They often directly aggregate vehicles with large differences in characteristics, resulting in poor consistency within the cluster and difficulty in accurately reflecting the actual regulation potential.

[0010] 2) Existing methods mostly use fixed expected departure power constraints, which do not fully distinguish between expected departure power, minimum guaranteed departure power and acceptable undercharging range authorized by the user. Therefore, it is difficult to effectively release the vehicle's adjustable space while ensuring the user's basic travel needs.

[0011] 3) While directly superimposing the vehicle's rated charging and discharging power is simple to calculate, it is easy to overlook the vehicle's online status, current battery level, and remaining dwell time, resulting in an overestimation of the cluster's adjustment capability; the precise feasible domain aggregation method also suffers from high model dimensionality and large computational load.

[0012] 4) A single power boundary can only describe the instantaneous adjustment capability of each time period and cannot constrain the cumulative power change across time periods. This may cause the charging task to be excessively postponed or the discharge schedule to be too large, making it difficult to ensure the feasibility of the energy trajectory throughout the entire scheduling cycle.

[0013] Therefore, it is necessary to propose an electric vehicle cluster regulation capability aggregation method based on dynamic off-site power boundary. Taking into account user psychological utility, this method comprehensively considers vehicle online status, current power, remaining dwell time, and off-site power demand to form a cluster charging and discharging power boundary and cumulative energy boundary that can be updated with vehicle status. This provides an accurate and executable regulation capability representation for electric vehicle cluster scheduling and virtual power plant operation control. Summary of the Invention

[0014] The purpose of this invention is to provide a method for aggregating the regulation capacity of electric vehicle clusters based on dynamic off-site power boundaries, in order to solve the problems of insufficient characterization of user behavior differences, fixed off-site power constraints, easy overestimation of cluster regulation capacity, and difficulty in guaranteeing cross-time period energy feasibility in the existing process of aggregating the regulation capacity of electric vehicle clusters. This provides an accurate, dynamic and executable regulation capacity boundary for electric vehicle clusters to participate in virtual power plant scheduling, demand response and vehicle-grid interaction.

[0015] In view of the shortcomings of existing technologies, the present invention focuses on solving the following technical problems:

[0016] 1. To address the issue that electric vehicle users have significant differences in access time, disconnection time, dwell time, and off-site power demand, making it difficult to accurately represent the adjustment characteristics of different users through unified aggregation; by constructing user behavior characteristics and performing clustering, vehicles with similar behavior patterns and adjustment needs are grouped into the same cluster, improving the consistency within the cluster and providing a reasonable vehicle grouping basis for subsequent adjustment capability aggregation.

[0017] 2. To address the problem that existing methods generally use fixed expected off-site charge constraints, which make it difficult to balance users' basic travel needs with the adjustable space of electric vehicles; by distinguishing between expected off-site charge and minimum guaranteed off-site charge, a dynamic off-site charge boundary relaxation coefficient is introduced to dynamically determine the lower boundary of the vehicle's off-site charge within the maximum acceptable undercharging range authorized by the user, thereby reasonably releasing the space for delayed charging and reverse discharge adjustment.

[0018] 3. To address the problem that directly superimposing the rated charging and discharging power of vehicles can easily overlook the vehicle's online status, current battery level, remaining dwell time, and off-site refueling demand, leading to an overestimation of the cluster's regulation capability; by calculating the dynamic acceptable battery level range of vehicles and the consistency rate of the state of charge demand progress, an equal-progress charging and discharging time scale is constructed, and based on this, the charging power boundary and discharging power boundary of the electric vehicle cluster are formed, improving the accuracy and feasibility of the cluster's instantaneous regulation capability characterization.

[0019] 4. To address the problem that using only a single-time-period power boundary is insufficient to constrain the vehicle's cross-time-period power trajectory, which may lead to excessive postponement of charging tasks and insufficient power supply at the departure point; by constructing an upper envelope and a lower envelope of cumulative energy for a single vehicle, and aggregating the cumulative energy envelopes of vehicles within the cluster, the upper and lower boundaries of the cumulative energy of the electric vehicle cluster are formed, thereby achieving a unified constraint on the range of cluster energy changes throughout the entire scheduling cycle.

[0020] To achieve the above objectives, this invention proposes a method for aggregating the regulation capacity of electric vehicle clusters based on dynamic off-site power boundaries, and provides the following technical solution:

[0021] The technical solution adopted by this invention to solve its technical problem is: a method for aggregating the regulation capacity of electric vehicle clusters based on dynamic off-site power boundaries, comprising the following steps:

[0022] S1: Obtain electric vehicle operating parameters and construct an EV scenario set reflecting the differences in behavior among various user types based on these parameters. Then, an electric vehicle charging behavior feature vector is constructed and standardized, and vehicle clusters are divided based on the standardized electric vehicle charging behavior feature vector.

[0023] S2: Calculate the expected departure charge and the minimum guaranteed departure charge. Based on the expected departure charge and the minimum guaranteed departure charge, determine the maximum acceptable undercharge amount authorized by the user and construct the dynamic departure charge boundary for a single vehicle. Construct the feasible range of vehicle departure charge based on the dynamic departure charge boundary for a single vehicle.

[0024] S3: Aggregate the charging and discharging power boundaries of vehicle clusters based on the consistency of state-of-charge demand schedule;

[0025] S4: Aggregate the upper and lower envelopes of energy of individual vehicles, construct the cumulative energy boundary of the vehicle cluster, and form the aggregation result of the regulation capacity of the electric vehicle cluster.

[0026] In step S1, the electric vehicle operating status parameters are obtained by first constructing an EV scenario set to reflect the differences in behavior among various types of users. In EV charging behavior scenarios Below, the operating state parameters of the i-th electric vehicle for:

[0027] ;

[0028] In the formula: and These refer to the vehicle's registration and departure times, respectively. The duration of vehicle stay; For the initial State of Charge (SOC) of the vehicle upon registration; For users who expect to be offline SOC; Allowing users to charge beyond their desired capacity; This indicates the minimum guaranteed state of charge at the station that the user can accept; and These are the minimum and maximum allowable SOC of the battery, respectively; This refers to the vehicle's battery capacity. and These are the maximum charging power and the maximum discharging power, respectively. and These are the charging and discharging efficiencies, respectively. A 0-1 variable representing whether a vehicle participates in the response;

[0029] Constructing and standardizing electric vehicle charging behavior feature vectors specifically includes:

[0030] Based on the periodicity of grid entry and exit times, sine-cosine cyclic encoding is performed on the grid entry and exit times. Combined with dwell time, initial state of charge, expected state of charge upon departure, minimum guaranteed state of charge upon departure, and permissible overcharge state of charge, the following feature vector of electric vehicle charging behavior is constructed:

[0031] ;

[0032] In the formula: and These refer to the vehicle's registration and departure times, respectively. The duration of vehicle stay; For the initial State of Charge (SOC) of the vehicle upon registration; For users who expect to be offline SOC; This indicates the minimum guaranteed state of charge at the station that the user can accept; Allowing users to charge beyond their desired capacity; Represents the i-th electric vehicle The following is a feature vector of charging behavior in the scenario; K represents the total number of scheduling periods contained in a scheduling cycle;

[0033] The different feature variables in the electric vehicle charging behavior feature vector are standardized, and the standardized complete charging behavior feature vector is represented as follows:

[0034] ;

[0035] ;

[0036] In the formula: This represents the standardized feature vector of the complete charging behavior. This represents the standardized d-th dimension of the charging behavior feature. This represents the d-th dimension of charging behavior characteristics before standardization. and These represent the mean and standard deviation of the d-th feature across all vehicle samples, respectively; d represents the feature dimension number. This represents a very small positive number that prevents the denominator from being zero; D is the total dimension of the eigenvectors.

[0037] The electric vehicle clusters described above are divided using the K-means clustering method based on standardized charging behavior feature vectors. The clustering objective function is:

[0038] ;

[0039] In the formula: C represents the set of electric vehicle cluster numbers; c represents the electric vehicle cluster number; Representing a scene The set of electric vehicles belonging to cluster c; Let i represent the standardized charging behavior feature vector of the i-th electric vehicle; Indicates the cluster center of cluster c; The second norm of a vector;

[0040] By solving the clustering objective function, electric vehicles with similar charging behavior characteristics are grouped into the same cluster, and the electric vehicle set corresponding to each electric vehicle cluster is obtained.

[0041] The specific calculations for the expected off-site power and the minimum guaranteed off-site power are as follows:

[0042] ;

[0043] ;

[0044] In the formula: Representing a scene The expected off-station charge of the i-th electric vehicle; Representing a scene The minimum guaranteed off-station battery power for the i-th electric vehicle; Indicates the desired state of charge when leaving the station; Indicates the minimum guaranteed off-site charged state; This represents the battery capacity of the i-th electric vehicle;

[0045] Based on the adjustable margin between the expected off-site charge and the minimum guaranteed off-site charge, determine the maximum acceptable undercharge amount authorized by the user in advance. :

[0046] .

[0047] The aforementioned dynamic departure power boundary for a single vehicle includes a lower boundary for dynamic departure power and an upper boundary for vehicle departure power.

[0048] For vehicle i belonging to electric vehicle cluster c, a dynamic off-station charge boundary relaxation coefficient is introduced to construct the lower boundary of the dynamic off-station charge. as follows:

[0049] ;

[0050] In the formula: Representing a scene The expected off-station charge of the i-th electric vehicle; This indicates the maximum acceptable undercharge amount authorized by the user in advance; , representing the dynamic off-site power boundary relaxation coefficient of cluster c, when At that time, the vehicle must reach the original expected departure charge level; when At that time, the vehicle can accept the maximum range of power reduction when leaving the station;

[0051] Based on the battery's maximum allowable state of charge and the user-allowed additional charging capacity, the upper boundary of the vehicle's off-site charge capacity is constructed as follows:

[0052] ; ;

[0053] In the formula: Representing a scene The upper boundary of the off-station battery level of the i-th electric vehicle; Indicates the maximum permissible state of charge of the battery; Indicates battery capacity; Indicates the expected off-site power consumption; This indicates that the user is allowed additional charging beyond the desired off-site charge level; This indicates that the user is allowed additional charging space beyond the expected off-site state of charge;

[0054] The feasible range of vehicle departure power is defined by the lower boundary of the dynamic departure power and the upper boundary of the departure power:

[0055] ;

[0056] In the formula: This indicates the time when the i-th electric vehicle leaves the network. The actual power consumption; and These represent the lower boundary and upper boundary of the dynamic off-site power, respectively.

[0057] Step S3, which aggregates the charging and discharging power boundaries of the vehicle cluster based on the consistency of state of charge demand schedule, includes the following steps:

[0058] S3.1 Determine the equivalent off-network time and the set of online response vehicles;

[0059] S3.2 Calculate the cut-off power demand and the dynamic acceptable power range;

[0060] S3.3 Consistency rate of demand schedule under calculated charge status;

[0061] S3.4 Calculate the charging and discharging timescale of the cluster at the same progress;

[0062] S3.5 aggregation forms the charging and discharging power boundary of the electric vehicle cluster;

[0063] Step S3.1 specifically includes:

[0064] Let the set of time periods within the current scheduling period be:

[0065] ;

[0066] In the formula: This represents the set of scheduling periods within the current scheduling period; K represents the total number of scheduling periods included in the current scheduling period.

[0067] Considering that some electric vehicles exhibit cross-scheduling cycle dwell behavior, let the i-th electric vehicle in the scenario be defined as... Extended off-network time below:

[0068] ;

[0069] In the formula: This represents the extended off-grid time of the i-th electric vehicle; and These represent the vehicle's entry time into the network and its original exit time, respectively.

[0070] The equivalent off-grid time for the vehicle within the current dispatch cycle is:

[0071] ;

[0072] In the formula: This represents the equivalent off-grid time of the i-th electric vehicle within the current scheduling cycle; Indicates the extended off-network time; K represents the total number of scheduling periods;

[0073] Define the scenario The set of vehicles in cluster c that are online and participating in the response during the next time period t:

[0074] ;

[0075] In the formula: Representing a scene The set of vehicles that are online and participating in the response within cluster c under t; Representing a scene The set of vehicles contained in cluster c; A 0-1 variable representing the online status of a vehicle; The variable represents the vehicle's participation in the response status (0-1); c represents the electric vehicle cluster number; t represents the scheduling period.

[0076] Vehicles that do not participate in the response operate in a plug-and-charge manner and are calculated separately as rigid charging loads, and do not participate in the subsequent cluster adjustable charging and discharging power boundary aggregation.

[0077] Step S3.2 specifically includes:

[0078] For vehicles that leave the network within the current scheduling cycle, the lower boundary of the dynamic off-site electricity consumption is taken as the required electricity consumption for the current cycle; for vehicles that stay across scheduling cycles, the minimum required electricity consumption at the end of the current cycle is calculated by back-calculating the remaining stay time in the next scheduling cycle, resulting in:

[0079] ;

[0080] In the formula: This represents the cut-off electricity demand of the i-th electric vehicle during the current scheduling cycle; Indicates the lower boundary of dynamic off-site power; Indicates the minimum permissible state of charge of the battery; Indicates battery capacity; Indicates charging efficiency; Indicates the maximum charging power; Indicates the extended off-network time; K represents the total number of scheduling periods; Indicates the duration of a single scheduling period;

[0081] Based on the cutoff demand for electricity and the remaining dwell time of the vehicle, calculate the dynamic minimum acceptable electricity level for the vehicle in time period t:

[0082] ;

[0083] In the formula: This represents the minimum acceptable energy level of the i-th electric vehicle during time period t; This indicates the cut-off electricity demand within the current scheduling cycle; Indicates charging efficiency; Indicates the maximum charging power; This represents the equivalent off-network time within the current scheduling cycle; Indicates the duration of a single scheduling period; Indicates the minimum permissible state of charge of the battery; Indicates battery capacity;

[0084] The maximum acceptable battery level for a vehicle during time period t is:

[0085] ;

[0086] In the formula: This represents the maximum acceptable battery level of the i-th electric vehicle during time period t; Indicates the upper limit of the vehicle's battery level when it leaves the station;

[0087] Therefore, the current battery level of the vehicle responding online should meet the following requirements:

[0088] ;

[0089] In the formula: This represents the current battery level of the i-th electric vehicle during time period t; and These represent the dynamic minimum acceptable battery level and the maximum acceptable battery level, respectively. This represents the set of vehicles that are online and participating in the response.

[0090] Step S3.3 specifically includes:

[0091] Calculate the vehicle's state of charge demand consistency rate based on its current battery level relative to the dynamically acceptable battery range:

[0092] ;

[0093] In the formula: This represents the consistency rate of the state of charge and demand progress of the i-th electric vehicle in time period t; Indicates the vehicle's current battery level; and These represent the vehicle's minimum and maximum acceptable battery level, respectively. This indicates a very small positive number that prevents the denominator from being zero;

[0094] The lower the state of charge demand progress consistency rate, the closer the vehicle is to the dynamic minimum acceptable charge level, and the more urgent its charging demand is; the higher the state of charge demand progress consistency rate, the higher the vehicle's charge margin, and the stronger its ability to provide discharge regulation.

[0095] Step S3.4 specifically includes:

[0096] Based on the dynamic acceptable battery capacity and maximum charging power of each online responding vehicle in the cluster, calculate the time scale for equal-progress charging of the cluster:

[0097] ;

[0098] In the formula: Indicates cluster c in scenario The isochronous charging timescale in the next time period t; This refers to the set of vehicles that are online and participating in the response; and These represent the vehicle's minimum and maximum acceptable battery level, respectively. Indicates vehicle charging efficiency; Indicates the vehicle's maximum charging power;

[0099] Based on the dynamic acceptable power range and maximum discharge power of each online response vehicle in the cluster, the time scale for equal-progress discharge of the cluster is calculated:

[0100] ;

[0101] In the formula: Indicates cluster c in scenario The isochronous discharge timescale in the next time period t; Indicates the vehicle's discharge efficiency; This indicates the vehicle's maximum discharge power.

[0102] Step S3.5 specifically includes:

[0103] Based on the equal-progress charging and discharging time scales of the electric vehicle cluster, the adjustable charging and discharging capabilities of online and participating vehicles within the cluster are aggregated to obtain the charging power boundary and discharging power boundary of the electric vehicle cluster:

[0104] ;

[0105] ;

[0106] In the formula: and Representing the scene respectively The charging power boundary and discharging power boundary of cluster c in time period t; This represents the set of vehicles within the cluster that are online and participating in the response. and These represent the vehicle's minimum and maximum acceptable battery level, respectively. and These represent the vehicle's charging efficiency and discharging efficiency, respectively. and These represent the time scales for equal-progress charging and equal-progress discharging of the cluster, respectively.

[0107] The actual charging and discharging power of the cluster must not exceed the corresponding power boundary, and simultaneous charging and discharging are not allowed within the same time period:

[0108] ;

[0109] ;

[0110] ;

[0111] In the formula: and These represent the actual charging power and discharging power of the cluster, respectively. and These are 0-1 variables representing the charging and discharging states of the cluster, respectively. and Representing the scene respectively The charging power boundary and discharging power boundary of cluster c in time period t;

[0112] In step S4, the upper and lower envelopes of the cumulative energy of a single vehicle are calculated in detail as follows;

[0113] The upper envelope of the cumulative net charging energy of the i-th electric vehicle up to time t is:

[0114] ;

[0115] In the formula: This represents the upper envelope of the cumulative net charging energy of the i-th electric vehicle up to time period t; Indicates the time of network access; Indicates the equivalent time of disconnection; Indicates battery capacity; Indicates the upper boundary of the off-site power quantity; Indicates the initial state of charge of the vehicle upon grid connection; Indicates vehicle charging efficiency; Indicates the vehicle's maximum charging power; Indicates the duration of a single scheduling period;

[0116] The lower envelope of the cumulative net charging energy of the i-th electric vehicle up to time t is:

[0117] ;

[0118] In the formula: This represents the lower envelope of the cumulative net charging energy of the i-th electric vehicle up to time period t; This represents the dynamic minimum acceptable power consumption for time period t; This indicates the cut-off electricity demand within the current scheduling cycle; Indicates the initial state of charge of the vehicle upon grid connection; Indicates the time of network access; Indicates the equivalent time of disconnection; Indicates battery capacity;

[0119] The specific calculation of the cluster's cumulative net charging energy in step S4 is as follows:

[0120] ;

[0121] In the formula: This represents the cumulative net charging energy of cluster c up to time period t; k represents the scheduling period number in the cumulative calculation. and These represent the charging power and discharging power of the cluster during time period k, respectively. and These represent the cluster's equivalent charging efficiency and equivalent discharging efficiency, respectively. Indicates the scheduling time interval.

[0122] The specific steps for calculating the cluster's cumulative net charging energy in step S4 are as follows;

[0123] By summing the upper and lower envelopes of the cumulative net charging energy of each vehicle within the same electric vehicle cluster, we obtain the lower and upper boundaries of the cluster's cumulative energy.

[0124] ;

[0125] ;

[0126] In the formula: and Representing the scene respectively The lower boundary and upper boundary of the cumulative energy of cluster c in time period t; This represents the set of electric vehicles participating in the response within cluster c; and These represent the lower envelope of the cumulative net charging energy and the upper envelope of the cumulative net charging energy for vehicle i up to time period t, respectively.

[0127] The cumulative net charging energy of the electric vehicle cluster meets:

[0128] ;

[0129] In the formula: Representing a scene The cumulative net charging energy of cluster c up to time period t.

[0130] The specific calculation of the aggregated results of the electric vehicle cluster regulation capability in step S4 is as follows;

[0131] The cluster charging power boundary, discharging power boundary, lower boundary of cumulative energy, and upper boundary of cumulative energy for each scheduling period are combined to form the electric vehicle cluster regulation capability boundary:

[0132] ;

[0133] In the formula: Representing a scene The set of adjustment capability boundaries for cluster c; and These represent the cluster charging power boundary and discharging power boundary for time period t, respectively. and These represent the lower boundary and upper boundary of the cluster's cumulative energy up to time period t, respectively. This represents the set of scheduling periods.

[0134] An electric vehicle cluster regulation capability aggregation system based on dynamic off-site power boundary, comprising:

[0135] The electric vehicle data processing and cluster partitioning module is used to obtain electric vehicle operating parameters, construct and standardize charging behavior feature vectors, and partition electric vehicle clusters.

[0136] The single-vehicle dynamic departure power boundary construction module is used to construct the single-vehicle dynamic departure power boundary based on the expected departure power, the minimum guaranteed departure power, and the maximum acceptable undercharge amount authorized by the user.

[0137] The vehicle cluster charging and discharging power boundary aggregation module is used to aggregate the vehicle cluster charging power boundary and discharging power boundary based on the vehicle online response status, dynamic acceptable power range and the consistency rate of state of charge demand progress.

[0138] The vehicle cluster cumulative energy boundary aggregation module is used to aggregate the upper and lower envelopes of the cumulative net charging energy of a single vehicle, obtain the upper and lower boundaries of the cumulative energy of the vehicle cluster, and form the aggregation result of the electric vehicle cluster regulation capability.

[0139] An electronic device includes a processor and a memory, the memory storing computer program instructions, the processor executing the instructions to implement the electric vehicle cluster regulation capability aggregation method based on dynamic off-site power boundary as described in any one of the preceding claims.

[0140] A computer-readable storage medium having a computer program stored thereon, which, when executed, implements a method for aggregating the regulation capabilities of electric vehicle clusters based on dynamic off-site power boundaries as described in any of the preceding claims.

[0141] By adopting the above technical solution, the present invention can produce the following beneficial effects:

[0142] 1. Improved the rationality of electric vehicle cluster division. This invention comprehensively considers user behavior characteristics such as vehicle access time, disconnection time, dwell time, initial state of charge, and off-site power demand, and groups vehicles with similar behavior patterns and adjustment needs into the same cluster, reducing differences within the cluster and improving the accuracy of subsequent adjustment capability aggregation results.

[0143] 2. Achieves dynamic representation of off-site power constraints. This invention distinguishes between the user's expected off-site power and the minimum guaranteed off-site power, and constructs a dynamic off-site power boundary by combining the user's maximum acceptable undercharging amount. While ensuring the user's basic travel needs, it releases adjustment space for vehicle delayed charging and reverse discharge, overcoming the problem of overly conservative fixed off-site power constraints.

[0144] 3. Balancing accuracy and computational efficiency of cluster power boundaries. This invention comprehensively considers vehicle online status, current battery level, remaining dwell time, and off-site refueling requirements. It forms cluster charging and discharging power boundaries by aggregating the consistency of state-of-charge demand progress and equal-progress charging and discharging time scales. This avoids overestimating the regulation capability caused by directly superimposing the vehicle's rated power, while also reducing the computational complexity of precise feasible region aggregation.

[0145] 4. Ensures the feasibility of cross-time period energy adjustment trajectory. This invention constructs upper and lower envelopes of the cumulative net charging energy of a single vehicle, and further aggregates them to form upper and lower boundaries of the cumulative energy of the electric vehicle cluster. This ensures that cluster scheduling not only meets the instantaneous power constraints of each time period, but also meets the cumulative energy constraints throughout the entire scheduling cycle, avoiding excessive postponement of charging tasks or excessive discharge scheduling.

[0146] 5. A unified, dynamic, and executable boundary for the regulation capability of electric vehicle clusters has been established. This invention unifies the expression of the cluster's charging and discharging power boundary and cumulative energy boundary, which can be dynamically updated according to changes in vehicle access status, power status, and off-site demand, providing a reliable boundary basis for electric vehicle clusters to participate in virtual power plant scheduling, demand response, and vehicle-to-grid interaction control. Attached Figure Description

[0147] Figure 1 This is a flowchart illustrating the overall process of the electric vehicle cluster regulation capability aggregation method involved in this invention.

[0148] Figure 2 This is a schematic diagram illustrating the variation of the sum of squared errors within each cluster under different numbers of clusters in an embodiment of the present invention;

[0149] Figure 3 This is a schematic diagram illustrating the construction of the dynamic departure power boundary for a single vehicle, as per the present invention.

[0150] Figure 4 This is a comparison diagram of the feasible domain of a single vehicle and the cluster adjustment capability boundary under different aggregation methods in an embodiment of the present invention. Detailed Implementation

[0151] The following will refer to the appendices in the embodiments of the present invention. Figure 1-4 The technical solutions in the embodiments of the present invention are clearly and completely described herein. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0152] This invention relates to a method for aggregating the regulation capacity of electric vehicle clusters based on dynamic off-site power boundaries, comprising the following steps:

[0153] S1: Obtain electric vehicle operating parameters and construct an EV scenario set reflecting the differences in behavior among various user types based on these parameters. Then, an electric vehicle charging behavior feature vector is constructed and standardized, and vehicle clusters are divided based on the standardized electric vehicle charging behavior feature vector.

[0154] S2: Calculate the expected departure charge and the minimum guaranteed departure charge. Based on the expected departure charge and the minimum guaranteed departure charge, determine the maximum acceptable undercharge amount authorized by the user and construct the dynamic departure charge boundary for a single vehicle. Construct the feasible range of vehicle departure charge based on the dynamic departure charge boundary for a single vehicle.

[0155] S3: Aggregate the charging and discharging power boundaries of vehicle clusters based on the consistency of state-of-charge demand schedule;

[0156] S4: Aggregate the upper and lower envelopes of energy of individual vehicles, construct the cumulative energy boundary of the vehicle cluster, and form the aggregation result of the regulation capacity of the electric vehicle cluster.

[0157] In step S1, the electric vehicle operating status parameters are obtained by first constructing an EV scenario set to reflect the differences in behavior among various types of users. In EV charging behavior scenarios Below, the operating state parameters of the i-th electric vehicle for:

[0158] ;

[0159] In the formula: and These refer to the vehicle's registration and departure times, respectively. The duration of vehicle stay; For the initial State of Charge (SOC) of the vehicle upon registration; For users who expect to be offline SOC; Allowing users to charge beyond their desired capacity; This indicates the minimum guaranteed state of charge at the station that the user can accept; and These are the minimum and maximum allowable SOC of the battery, respectively; This refers to the vehicle's battery capacity. and These are the maximum charging power and the maximum discharging power, respectively. and These are the charging and discharging efficiencies, respectively. A 0-1 variable representing whether a vehicle participates in the response;

[0160] Constructing and standardizing electric vehicle charging behavior feature vectors specifically includes:

[0161] Based on the periodicity of grid entry and exit times, sine-cosine cyclic encoding is performed on the grid entry and exit times. Combined with dwell time, initial state of charge, expected state of charge upon departure, minimum guaranteed state of charge upon departure, and permissible overcharge state of charge, the following feature vector of electric vehicle charging behavior is constructed:

[0162] ;

[0163] In the formula: and These refer to the vehicle's registration and departure times, respectively. The duration of vehicle stay; For the initial State of Charge (SOC) of the vehicle upon registration; For users who expect to be offline SOC; This indicates the minimum guaranteed state of charge at the station that the user can accept; Allowing users to charge beyond their desired capacity; Represents the i-th electric vehicle The following is a feature vector of charging behavior in the scenario; K represents the total number of scheduling periods contained in a scheduling cycle;

[0164] The different feature variables in the electric vehicle charging behavior feature vector are standardized, and the standardized complete charging behavior feature vector is represented as follows:

[0165] ;

[0166] ;

[0167] In the formula: This represents the standardized feature vector of the complete charging behavior. This represents the standardized d-th dimension of the charging behavior feature. This represents the d-th dimension of charging behavior characteristics before standardization. and These represent the mean and standard deviation of the d-th feature across all vehicle samples, respectively; d represents the feature dimension number. This represents a very small positive number that prevents the denominator from being zero; D is the total dimension of the eigenvectors.

[0168] The electric vehicle clusters described above are divided using the K-means clustering method based on standardized charging behavior feature vectors. The clustering objective function is:

[0169] ;

[0170] In the formula: C represents the set of electric vehicle cluster numbers; c represents the electric vehicle cluster number; Representing a scene The set of electric vehicles belonging to cluster c; Let i represent the standardized charging behavior feature vector of the i-th electric vehicle; Indicates the cluster center of cluster c; The second norm of a vector;

[0171] By solving the clustering objective function, electric vehicles with similar charging behavior characteristics are grouped into the same cluster, and the electric vehicle set corresponding to each electric vehicle cluster is obtained.

[0172] The specific calculations for the expected off-site power and the minimum guaranteed off-site power are as follows:

[0173] ;

[0174] ;

[0175] In the formula: Representing a scene The expected off-station charge of the i-th electric vehicle; Representing a scene The minimum guaranteed off-station battery power for the i-th electric vehicle; Indicates the desired state of charge when leaving the station; Indicates the minimum guaranteed off-site charged state; This represents the battery capacity of the i-th electric vehicle;

[0176] Based on the adjustable margin between the expected off-site charge and the minimum guaranteed off-site charge, determine the maximum acceptable undercharge amount authorized by the user in advance. :

[0177] .

[0178] The aforementioned dynamic departure power boundary for a single vehicle includes a lower boundary for dynamic departure power and an upper boundary for vehicle departure power.

[0179] For vehicle i belonging to electric vehicle cluster c, a dynamic off-station charge boundary relaxation coefficient is introduced to construct the lower boundary of the dynamic off-station charge. as follows:

[0180] ;

[0181] In the formula: Representing a scene The expected off-station charge of the i-th electric vehicle; This indicates the maximum acceptable undercharge amount authorized by the user in advance; , representing the dynamic off-site power boundary relaxation coefficient of cluster c, when At that time, the vehicle must reach the original expected departure charge level; when At that time, the vehicle can accept the maximum range of power reduction when leaving the station;

[0182] Based on the battery's maximum allowable state of charge and the user-allowed additional charging capacity, the upper boundary of the vehicle's off-site charge capacity is constructed as follows:

[0183] ; ;

[0184] In the formula: Representing a scene The upper boundary of the off-station battery level of the i-th electric vehicle; Indicates the maximum permissible state of charge of the battery; Indicates battery capacity; Indicates the expected off-site power consumption; This indicates that the user is allowed additional charging beyond the desired off-site charge level; This indicates that the user is allowed additional charging space beyond the expected off-site state of charge;

[0185] The feasible range of vehicle departure power is defined by the lower boundary of the dynamic departure power and the upper boundary of the departure power:

[0186] ;

[0187] In the formula: This indicates the time when the i-th electric vehicle leaves the network. The actual power consumption; and These represent the lower boundary and upper boundary of the dynamic off-site power, respectively.

[0188] Step S3, which aggregates the charging and discharging power boundaries of the vehicle cluster based on the consistency of state of charge demand schedule, includes the following steps:

[0189] S3.1 Determine the equivalent off-network time and the set of online response vehicles;

[0190] S3.2 Calculate the cut-off power demand and the dynamic acceptable power range;

[0191] S3.3 Consistency rate of demand schedule under calculated charge status;

[0192] S3.4 Calculate the charging and discharging timescale of the cluster at the same progress;

[0193] S3.5 aggregation forms the charging and discharging power boundary of the electric vehicle cluster;

[0194] Step S3.1 specifically includes:

[0195] Let the set of time periods within the current scheduling period be:

[0196] ;

[0197] In the formula: This represents the set of scheduling periods within the current scheduling period; K represents the total number of scheduling periods included in the current scheduling period.

[0198] Considering that some electric vehicles exhibit cross-scheduling cycle dwell behavior, let the i-th electric vehicle in the scenario be defined as... Extended off-network time below:

[0199] ;

[0200] In the formula: This represents the extended off-grid time of the i-th electric vehicle; and These represent the vehicle's entry time into the network and its original exit time, respectively.

[0201] The equivalent off-grid time for the vehicle within the current dispatch cycle is:

[0202] ;

[0203] In the formula: This represents the equivalent off-grid time of the i-th electric vehicle within the current scheduling cycle; Indicates the extended off-network time; K represents the total number of scheduling periods;

[0204] Define the scenario The set of vehicles in cluster c that are online and participating in the response during the next time period t:

[0205] ;

[0206] In the formula: Representing a scene The set of vehicles that are online and participating in the response within cluster c under t; Representing a scene The set of vehicles contained in cluster c; A 0-1 variable representing the online status of a vehicle; The variable represents the vehicle's participation in the response status (0-1); c represents the electric vehicle cluster number; t represents the scheduling period.

[0207] Vehicles that do not participate in the response operate in a plug-and-charge manner and are calculated separately as rigid charging loads, and do not participate in the subsequent cluster adjustable charging and discharging power boundary aggregation.

[0208] Step S3.2 specifically includes:

[0209] For vehicles that leave the network within the current scheduling cycle, the lower boundary of the dynamic off-site electricity consumption is taken as the required electricity consumption for the current cycle; for vehicles that stay across scheduling cycles, the minimum required electricity consumption at the end of the current cycle is calculated by back-calculating the remaining stay time in the next scheduling cycle, resulting in:

[0210] ;

[0211] In the formula: This represents the cut-off electricity demand of the i-th electric vehicle during the current scheduling cycle; Indicates the lower boundary of dynamic off-site power; Indicates the minimum permissible state of charge of the battery; Indicates battery capacity; Indicates charging efficiency; Indicates the maximum charging power; Indicates the extended off-network time; K represents the total number of scheduling periods; Indicates the duration of a single scheduling period;

[0212] Based on the cutoff demand for electricity and the remaining dwell time of the vehicle, calculate the dynamic minimum acceptable electricity level for the vehicle in time period t:

[0213] ;

[0214] In the formula: This represents the minimum acceptable energy level of the i-th electric vehicle during time period t; This indicates the cut-off electricity demand within the current scheduling cycle; Indicates charging efficiency; Indicates the maximum charging power; This represents the equivalent off-network time within the current scheduling cycle; Indicates the duration of a single scheduling period; Indicates the minimum permissible state of charge of the battery; Indicates battery capacity;

[0215] The maximum acceptable battery level for a vehicle during time period t is:

[0216] ;

[0217] In the formula: This represents the maximum acceptable battery level of the i-th electric vehicle during time period t; Indicates the upper limit of the vehicle's battery level when it leaves the station;

[0218] Therefore, the current battery level of the vehicle responding online should meet the following requirements:

[0219] ;

[0220] In the formula: This represents the current battery level of the i-th electric vehicle during time period t; and These represent the dynamic minimum acceptable battery level and the maximum acceptable battery level, respectively. This represents the set of vehicles that are online and participating in the response.

[0221] Step S3.3 specifically includes:

[0222] Calculate the vehicle's state of charge demand consistency rate based on its current battery level relative to the dynamically acceptable battery range:

[0223] ;

[0224] In the formula: This represents the consistency rate of the state of charge and demand progress of the i-th electric vehicle in time period t; Indicates the vehicle's current battery level; and These represent the vehicle's minimum and maximum acceptable battery level, respectively. This indicates a very small positive number that prevents the denominator from being zero;

[0225] The lower the state of charge demand progress consistency rate, the closer the vehicle is to the dynamic minimum acceptable charge level, and the more urgent its charging demand is; the higher the state of charge demand progress consistency rate, the higher the vehicle's charge margin, and the stronger its ability to provide discharge regulation.

[0226] Step S3.4 specifically includes:

[0227] Based on the dynamic acceptable battery capacity and maximum charging power of each online responding vehicle in the cluster, calculate the time scale for equal-progress charging of the cluster:

[0228] ;

[0229] In the formula: Indicates cluster c in scenario The isochronous charging timescale in the next time period t; This refers to the set of vehicles that are online and participating in the response; and These represent the vehicle's minimum and maximum acceptable battery level, respectively. Indicates vehicle charging efficiency; Indicates the vehicle's maximum charging power;

[0230] Based on the dynamic acceptable power range and maximum discharge power of each online response vehicle in the cluster, the time scale for equal-progress discharge of the cluster is calculated:

[0231] ;

[0232] In the formula: Indicates cluster c in scenario The isochronous discharge timescale in the next time period t; Indicates the vehicle's discharge efficiency; This indicates the vehicle's maximum discharge power.

[0233] Step S3.5 specifically includes:

[0234] Based on the equal-progress charging and discharging time scales of the electric vehicle cluster, the adjustable charging and discharging capabilities of online and participating vehicles within the cluster are aggregated to obtain the charging power boundary and discharging power boundary of the electric vehicle cluster:

[0235] ;

[0236] ;

[0237] In the formula: and Representing the scene respectively The charging power boundary and discharging power boundary of cluster c in time period t; This represents the set of vehicles within the cluster that are online and participating in the response. and These represent the vehicle's minimum and maximum acceptable battery level, respectively. and These represent the vehicle's charging efficiency and discharging efficiency, respectively. and These represent the time scales for equal-progress charging and equal-progress discharging of the cluster, respectively.

[0238] The actual charging and discharging power of the cluster must not exceed the corresponding power boundary, and simultaneous charging and discharging are not allowed within the same time period:

[0239] ;

[0240] ;

[0241] ;

[0242] In the formula: and These represent the actual charging power and discharging power of the cluster, respectively. and These are 0-1 variables representing the charging and discharging states of the cluster, respectively.

[0243] In step S4, the upper and lower envelopes of the cumulative energy of a single vehicle are calculated in detail as follows;

[0244] The upper envelope of the cumulative net charging energy of the i-th electric vehicle up to time t is:

[0245] ;

[0246] In the formula: This represents the upper envelope of the cumulative net charging energy of the i-th electric vehicle up to time period t; Indicates the time of network access; Indicates the equivalent time of disconnection; Indicates battery capacity; Indicates the upper boundary of the off-site power quantity; Indicates the initial state of charge of the vehicle upon grid connection; Indicates vehicle charging efficiency; Indicates the vehicle's maximum charging power; Indicates the duration of a single scheduling period;

[0247] The lower envelope of the cumulative net charging energy of the i-th electric vehicle up to time t is:

[0248] ;

[0249] In the formula: This represents the lower envelope of the cumulative net charging energy of the i-th electric vehicle up to time period t; This represents the dynamic minimum acceptable power consumption for time period t; This indicates the cut-off electricity demand within the current scheduling cycle; Indicates the initial state of charge of the vehicle upon grid connection; Indicates the time of network access; Indicates the equivalent time of disconnection; Indicates battery capacity;

[0250] The specific calculation of the cluster's cumulative net charging energy in step S4 is as follows:

[0251] ;

[0252] In the formula: This represents the cumulative net charging energy of cluster c up to time period t; k represents the scheduling period number in the cumulative calculation. and These represent the charging power and discharging power of the cluster during time period k, respectively. and These represent the cluster's equivalent charging efficiency and equivalent discharging efficiency, respectively. Indicates the scheduling time interval.

[0253] The specific steps for calculating the cluster's cumulative net charging energy in step S4 are as follows;

[0254] By summing the upper and lower envelopes of the cumulative net charging energy of each vehicle within the same electric vehicle cluster, we obtain the lower and upper boundaries of the cluster's cumulative energy.

[0255] ;

[0256] ;

[0257] In the formula: and Representing the scene respectively The lower boundary and upper boundary of the cumulative energy of cluster c in time period t; This represents the set of electric vehicles participating in the response within cluster c; and These represent the lower envelope of the cumulative net charging energy and the upper envelope of the cumulative net charging energy for vehicle i up to time period t, respectively.

[0258] The cumulative net charging energy of the electric vehicle cluster meets:

[0259] ;

[0260] In the formula: Representing a scene The cumulative net charging energy of cluster c up to time period t.

[0261] The specific calculation of the aggregated results of the electric vehicle cluster regulation capability in step S4 is as follows;

[0262] The cluster charging power boundary, discharging power boundary, lower boundary of cumulative energy, and upper boundary of cumulative energy for each scheduling period are combined to form the electric vehicle cluster regulation capability boundary:

[0263] ;

[0264] In the formula: Representing a scene The set of adjustment capability boundaries for cluster c; and These represent the cluster charging power boundary and discharging power boundary for time period t, respectively. and These represent the lower boundary and upper boundary of the cluster's cumulative energy up to time period t, respectively. This represents the set of scheduling periods.

[0265] The following is a specific example to illustrate this point:

[0266] Aggregation process of cluster regulation capability based on a virtual power plant scenario of electric vehicles in a certain province;

[0267] This embodiment takes electric vehicles aggregated in a virtual power plant of electric vehicle type in a certain province as the research object, sets the scheduling cycle to 24 hours, the scheduling time interval to 15 minutes, and each scheduling cycle includes 96 scheduling time periods. Through electric vehicle user behavior clustering, construction of single vehicle dynamic off-site power boundary, aggregation of cluster charging and discharging power boundary, and construction of cluster cumulative energy boundary, the electric vehicle cluster regulation capability boundary is formed.

[0268] 1. Electric vehicle user behavior clustering

[0269] First, user behavior data for each electric vehicle is acquired, including its network entry time, network exit time, dwell time, initial state of charge (SOC), expected SOC upon departure, minimum guaranteed SOC upon departure, and permissible overcharge SOC. Technical parameters such as vehicle battery capacity, maximum charging / discharging power, charging / discharging efficiency, and safe SOC range are also obtained. To address the periodicity of network entry and exit times, cyclic encoding using sine and cosine functions is employed, and each user behavior variable is standardized. Subsequently, K-means clustering is performed, and the number of clusters is determined based on in-cluster error sum of squares and silhouette coefficient. This embodiment ultimately categorizes electric vehicles into five clusters: daytime office park vehicles, public short-term fast-charging vehicles, afternoon commercial flexible vehicles, ride-hailing vehicles, and nighttime residential slow-charging vehicles, providing a basis for subsequent dynamic boundary construction.

[0270] 2. Construction of dynamic departure power boundary for single vehicle

[0271] Based on the vehicle's battery capacity, expected state of charge (SBC) at departure, and minimum guaranteed SBC at departure, the expected and minimum guaranteed SBC at departure are calculated separately. The maximum acceptable undercharge is determined based on user-reported or authorized information. A dynamic SBC at departure boundary relaxation coefficient, ranging from 0 to 1, is set for different vehicle clusters. Within the user-authorized range, the lower boundary of the vehicle's dynamic SBC at departure is determined. When the relaxation coefficient is 0, the lower boundary equals the expected SBC at departure. As the relaxation coefficient increases, the available undercharge space increases accordingly, but the lower boundary must not be lower than the minimum guaranteed SBC at departure. Simultaneously, the upper boundary of the SBC at departure is determined based on the expected SBC at departure, the allowed overcharge, and the upper limit of the battery SBC at departure safety, thus forming a feasible range for the dynamic SBC at departure for a single vehicle.

[0272] 3. Boundary aggregation of charging and discharging power of electric vehicle clusters

[0273] Based on the vehicle's entry and exit times and response status, the set of vehicles online and participating in response within each scheduling period is determined, while vehicles not participating in response are treated separately as rigid charging loads. For vehicles residing across scheduling cycles, a cycle-end truncation method is used to determine their current cycle's required electricity. Furthermore, combining the vehicle's minimum safe electricity level, departure electricity boundary, remaining dwell time, and maximum charging capacity, the dynamic minimum acceptable electricity level and maximum acceptable electricity level for each period are calculated. The state-of-charge demand consistency rate is calculated based on the vehicle's current electricity level relative to the acceptable electricity level range, and this is used to determine the cluster equal-progress charging time scale and equal-progress discharging time scale. Finally, these are aggregated to form the cluster charging power boundary and discharging power boundary for each scheduling period, thereby avoiding overestimation of regulation capacity caused by directly adding the vehicle's rated power.

[0274] 4. Construction of the cumulative energy boundary of electric vehicle clusters

[0275] To ensure the feasibility of cross-time period power trajectory, the cumulative net charging energy of a single vehicle is calculated based on the vehicle's charging and discharging power, charging and discharging efficiency, and scheduling time interval. Combined with the vehicle's initial power level, dynamic minimum acceptable power level, maximum acceptable power level, remaining dwell time, and maximum charging capacity, a lower envelope and an upper envelope of cumulative energy for each vehicle are constructed. The lower envelope prevents excessive postponement of charging tasks, while the upper envelope prevents vehicle power levels from exceeding permissible limits. Subsequently, the upper and lower envelopes of cumulative energy for each participating vehicle within the same cluster are summed to obtain the upper and lower boundaries of the cluster's cumulative energy. These boundaries, along with the cluster's charging and discharging power boundary, form the set of electric vehicle cluster regulation capability boundaries.

[0276] To demonstrate the effectiveness and superiority of the proposed electric vehicle cluster regulation capability aggregation method based on dynamic off-site power boundaries, a case study analysis can be conducted using actual data, and combined with the appendix... Figure 1 —4. Analyze the results.

[0277] Figure 1 This is a flowchart illustrating the overall process of the electric vehicle cluster regulation capability aggregation method involved in this invention. Figure 1 As can be seen, this invention first acquires electric vehicle user behavior data and divides it into user behavior clusters. Then, it constructs a dynamic off-site power boundary for each vehicle, and based on this, sequentially forms a cluster charging / discharging power boundary and a cluster cumulative energy boundary, ultimately obtaining a set of electric vehicle cluster regulation capability boundaries. Each step has a clear input-output relationship, forming a complete process of progressively aggregating individual vehicle behavior information to cluster regulation capability boundaries.

[0278] Figure 2 This diagram illustrates the variation of the sum of squared intra-cluster errors under different numbers of clusters in an embodiment of the present invention. Figure 2 As can be seen, the sum of squared errors within clusters continuously decreases with the increase in the number of clusters, indicating that increasing the number of clusters can improve the consistency of user behavior characteristics within the cluster. When the number of clusters reaches 5, the rate of decrease in the sum of squared errors within clusters slows down significantly, and the improvement in clustering accuracy brought by further increasing the number of clusters is limited. Therefore, this embodiment sets the number of clusters to 5, which can achieve a better balance between consistency within the cluster and clustering complexity.

[0279] The K-means algorithm was used to cluster electric vehicle users, resulting in five typical user groups. The specific cluster center parameters are shown in Table 1. The proportions of the five user groups were 20.98%, 15.78%, 18.64%, 10.16%, and 34.44%, respectively, which can reflect the differences in user access patterns, energy replenishment needs, and adjustment potential under different charging scenarios.

[0280] Table 1. Behavioral characteristic parameters of typical users (cluster centers) in this embodiment.

[0281]

[0282] As shown in Table 1, Category 1 office park daytime users typically connect to the network at 08:24 and disconnect at 16:45, with an average stay of 8.36 hours. They have a relatively long and stable daytime adjustment window, allowing them to participate in charging sequence adjustments while meeting their off-site power demand. Category 2 public short-term fast-charging users have an average stay of only 2.48 hours, with a connected charge level of 19.45%, while expecting an off-site charge level of 76.33%. Their charging time is short and their power demand is concentrated, resulting in relatively limited adjustment capabilities. Category 3 afternoon flexible commercial users mainly connect in the afternoon, with an average stay of 3.17 hours. They have a relatively high connected charge level and some charging transfer space, but are still limited by their short stay time. Category 4 ride-hailing vehicle users typically charge in the morning, with an average stay of 2.61 hours. They have the lowest connected charge level at only 16.79%, and expect an off-site charge level of 80.73%, indicating that their charging demand is rigid and their available adjustment space is small. Category 5, nighttime slow-charging users, account for the highest proportion at 34.44%. They typically connect to the network at 19:19 and disconnect at 07:17 the next day, with an average dwell time of 11.96 hours, providing the longest window for charging and regulation. At the same time, their allowable overcharge state of charge is 5.44%, and under the condition of ensuring the minimum off-site guaranteed state of charge, they have significant potential for delayed charging and V2G regulation.

[0283] The above results indicate significant differences among different user types in terms of access time, dwell time, initial charge, and off-site charge requirements. Specifically, public short-term fast-charging users and ride-hailing users have more urgent charging needs, while office park daytime users and nighttime residential slow-charging users have longer dwell times and greater flexibility in timing adjustments. Therefore, grouping vehicles with similar behavioral characteristics into the same cluster can provide a classification basis for subsequently constructing dynamic off-site charge boundaries and aggregated charge / discharge power boundaries by cluster.

[0284] Figure 3 This is a schematic diagram illustrating the construction of the dynamic departure power boundary for a single vehicle, as per the present invention. Figure 3 It is evident that there is a certain adjustable range between the user's desired departure charge and the minimum guaranteed departure charge. This invention dynamically adjusts the lower boundary of the departure charge within the user-authorized maximum acceptable undercharging range by setting a dynamic departure charge boundary relaxation coefficient. As the relaxation coefficient increases, the adjustment space for delayed charging and reverse discharge of the vehicle expands accordingly; simultaneously, the dynamic lower boundary of the departure charge never falls below the minimum guaranteed departure charge, ensuring the user's basic travel needs are met.

[0285] Figure 4This is a comparison diagram of the feasible region of a single vehicle and the boundary of cluster adjustment capability under different aggregation methods in an embodiment of the present invention. To intuitively illustrate the differences in the characterization of the feasible region of electric vehicles by different aggregation methods, taking two electric vehicles and three consecutive scheduling periods as an example, the feasible region of a single vehicle and the aggregated feasible region are represented as three-dimensional polyhedra. Figure 4 It is evident that the feasible regions for individual vehicles of EV1 and EV2 are not simple power rectangles, but rather irregular polyhedra constrained by the upper limits of individual vehicle charging and discharging power, current battery status, and off-site battery demand. This indicates that the power regulation capability that a vehicle can provide at a certain time period is also affected by the energy availability in subsequent time periods. If the upper limits of power and energy of each vehicle are simply added together, it is easy to include power-energy combinations that cannot be simultaneously achieved at the individual vehicle level into the cluster boundary, thus leading to an overestimation of the regulation capability.

[0286] Comparing the results of the cluster adjustment capability boundary, compared with the precise aggregation method based on Minkowski sums, the boundary formed by the method of this invention can better cover the main adjustable region of the cluster, with only some approximate deviations at certain boundary locations. Calculated according to the dimension-normalized volume ratio, the adjustment capability loss of the method of this invention relative to the precise aggregation boundary is 6.63%, indicating that it can still accurately characterize the main power-energy adjustment space of the electric vehicle cluster without performing precise summation of high-dimensional polyhedra.

[0287] Table 2 Comparison of aggregation method results under different vehicle sizes in this embodiment.

[0288]

[0289] Table 2 compares the results of the proposed method with those of the Minkowski and exact aggregation methods under different vehicle scales. In small-scale scenarios, the total cost of the final solutions obtained by the two methods is quite similar; when the number of vehicles is 10, the difference is only 0.80 yuan. When the number of vehicles is 50 and 100, the total cost of the proposed method is 1.80% and 4.74% higher than that of the Minkowski and exact aggregation methods, respectively, indicating that the aggregation boundary formed by the proposed method has a certain degree of conservatism, but still retains the main adjustable space of the cluster. As the number of vehicles increases, the computation time of the Minkowski and exact aggregation methods increases rapidly. When the number of vehicles increases from 100 to 200, the computation time increases from 147.92s to 3910.04s, which is about 29.7 times the computation time of the proposed method; when the number of vehicles reaches 500, the Minkowski and exact aggregation methods still have not obtained results after more than 12 hours, while the proposed method can complete the calculation within 268.10s. Therefore, the method of the present invention significantly reduces the computational complexity of large-scale electric vehicle cluster regulation capacity aggregation with a smaller boundary approximation loss and a certain degree of economic conservatism, and is more suitable for rapid evaluation and boundary update of the regulation capacity of virtual power plants with a large number of vehicles.

[0290] The above results show that the present invention can make full use of the charging and discharging regulation space of electric vehicle clusters while ensuring the minimum off-site power demand of users; at the same time, while reducing the computational complexity of large-scale vehicle aggregation, it avoids the overestimation of regulation capability caused by the simple superposition of power and energy boundaries, and provides accurate and executable cluster regulation capability boundaries for scheduling planning, demand response and vehicle-grid interaction control of electric vehicle virtual power plants.

[0291] This invention also relates to an electric vehicle cluster regulation capability aggregation system based on dynamic off-site power boundary, comprising:

[0292] The electric vehicle data processing and cluster partitioning module is used to obtain electric vehicle operating parameters, construct and standardize charging behavior feature vectors, and partition electric vehicle clusters.

[0293] The single-vehicle dynamic departure power boundary construction module is used to construct the single-vehicle dynamic departure power boundary based on the expected departure power, the minimum guaranteed departure power, and the maximum acceptable undercharge amount authorized by the user.

[0294] The vehicle cluster charging and discharging power boundary aggregation module is used to aggregate the vehicle cluster charging power boundary and discharging power boundary based on the vehicle online response status, dynamic acceptable power range and the consistency rate of state of charge demand progress.

[0295] The vehicle cluster cumulative energy boundary aggregation module is used to aggregate the upper and lower envelopes of the cumulative net charging energy of a single vehicle, obtain the upper and lower boundaries of the cumulative energy of the vehicle cluster, and form the aggregation result of the electric vehicle cluster regulation capability.

[0296] The present invention also includes an electronic device comprising a processor and a memory, the memory for storing computer program instructions, the processor executing the instructions to implement the electric vehicle cluster regulation capability aggregation method based on dynamic off-site power boundary as described in any of the foregoing.

[0297] The present invention also relates to a computer-readable storage medium having a computer program stored thereon, which, when executed, can implement the electric vehicle cluster regulation capability aggregation method based on dynamic off-site power boundary as described in any of the foregoing.

Claims

1. A method for aggregating the regulation capacity of electric vehicle clusters based on dynamic off-site power boundaries, characterized in that, Includes the following steps: S1: Obtain electric vehicle operating parameters and construct an EV scenario set reflecting the differences in behavior among various user types based on these parameters. Then, an electric vehicle charging behavior feature vector is constructed and standardized, and vehicle clusters are divided based on the standardized electric vehicle charging behavior feature vector. S2: Calculate the expected departure charge and the minimum guaranteed departure charge. Based on the expected departure charge and the minimum guaranteed departure charge, determine the maximum acceptable undercharge amount authorized by the user and construct the dynamic departure charge boundary for a single vehicle. Construct the feasible range of vehicle departure charge based on the dynamic departure charge boundary for a single vehicle. S3: Aggregate the charging and discharging power boundaries of vehicle clusters based on the consistency of state-of-charge demand schedule; S4: Aggregate the upper and lower envelopes of energy of a single vehicle to construct the cumulative energy boundary of the vehicle cluster and form the aggregation result of the regulation capability of the electric vehicle cluster; Step S3, which aggregates the charging and discharging power boundaries of the vehicle cluster based on the consistency of state of charge demand schedule, includes the following steps: S3.1 Determine the equivalent off-network time and the set of online response vehicles; S3.2 Calculate the cut-off power demand and the dynamic acceptable power range; S3.3 Calculate the consistency rate of the state of charge requirement schedule; S3.4 Calculate the charging and discharging timescale of the cluster at the same progress; S3.5 aggregation forms the charging and discharging power boundary of the electric vehicle cluster; Step S3.1 specifically includes: Let the set of time periods within the current scheduling period be: ; In the formula: This represents the set of scheduling periods within the current scheduling period; K represents the total number of scheduling periods included in the current scheduling period. Considering that some electric vehicles exhibit cross-scheduling cycle dwell behavior, let the i-th electric vehicle in the scenario be defined as... Extended off-network time below: ; In the formula: This represents the extended off-grid time of the i-th electric vehicle; and These represent the vehicle's entry time and original exit time, respectively; K represents the total number of scheduling periods included in the current scheduling cycle. The equivalent off-grid time for the vehicle within the current dispatch cycle is: ; In the formula: This represents the equivalent off-grid time of the i-th electric vehicle within the current scheduling cycle; Indicates the extended off-network time; K represents the total number of scheduling periods; Define the scenario The set of vehicles online and participating in the response in cluster c during the next time period t: ; In the formula: Representing a scene The set of vehicles that are online and participating in the response within cluster c under t; Representing a scene The set of vehicles contained in cluster c; A 0-1 variable representing the online status of a vehicle; 0-1 variables represent the vehicle's participation in the response; c represents the electric vehicle cluster number; t represents the scheduling period. Vehicles that do not participate in the response operate in a plug-and-charge manner and are calculated separately as rigid charging loads, and do not participate in the subsequent cluster adjustable charging and discharging power boundary aggregation; Step S3.4 specifically includes: Based on the dynamic acceptable battery capacity and maximum charging power of each online responding vehicle in the cluster, calculate the time scale for equal-progress charging of the cluster: ; In the formula: Indicates cluster c in scenario The isochronous charging timescale in the next time period t; This refers to the set of vehicles that are online and participating in the response; and These represent the vehicle's minimum and maximum acceptable battery level, respectively. Indicates vehicle charging efficiency; Indicates the vehicle's maximum charging power; Based on the dynamic acceptable power range and maximum discharge power of each online response vehicle in the cluster, the time scale for equal-progress discharge of the cluster is calculated: ; In the formula: Indicates cluster c in scenario The isochronous discharge timescale in the next time period t; Indicates the vehicle's discharge efficiency; This indicates the vehicle's maximum discharge power.

2. The method for aggregating the regulation capability of electric vehicle clusters based on dynamic off-site power boundaries according to claim 1, characterized in that, In step S1, the electric vehicle operating status parameters are obtained by first constructing an EV scenario set to reflect the differences in behavior among various types of users. In EV charging behavior scenarios Below, the operating state parameters of the i-th electric vehicle for: ; In the formula: and These refer to the vehicle's registration and deregistration times, respectively. Duration of vehicle stay; For the initial State of Charge (SOC) of the vehicle upon registration; For users who expect to be offline SOC; Allowing users to charge beyond their desired capacity; This indicates the minimum guaranteed state of charge at the station that the user can accept; and These are the minimum and maximum allowable SOC of the battery, respectively; This refers to the vehicle's battery capacity. and These are the maximum charging power and the maximum discharging power, respectively. and These are the charging and discharging efficiencies, respectively. This is a 0-1 variable representing whether a vehicle participates in the response.

3. The method for aggregating the regulation capability of electric vehicle clusters based on dynamic off-site power boundaries according to claim 2, characterized in that, Constructing and standardizing electric vehicle charging behavior feature vectors specifically includes: Based on the periodicity of grid entry and exit times, sine-cosine cyclic encoding is performed on the grid entry and exit times. Combined with dwell time, initial state of charge, expected state of charge upon departure, minimum guaranteed state of charge upon departure, and permissible overcharge state of charge, the following feature vector of electric vehicle charging behavior is constructed: ; In the formula: and These refer to the vehicle's registration and deregistration times, respectively. Duration of vehicle stay; For the initial State of Charge (SOC) of the vehicle upon registration; For users who expect to be offline SOC; This indicates the minimum guaranteed state of charge at the station that the user can accept; Allowing users to charge beyond their desired capacity; Represents the i-th electric vehicle The following is a feature vector of charging behavior in the scenario; K represents the total number of scheduling periods contained in a scheduling cycle; The different feature variables in the electric vehicle charging behavior feature vector are standardized, and the standardized complete charging behavior feature vector is represented as follows: ; ; In the formula: This represents the standardized feature vector of the complete charging behavior. This represents the standardized d-th dimension of the charging behavior feature. This represents the d-th dimension of charging behavior characteristics before standardization. and These represent the mean and standard deviation of the d-th feature across all vehicle samples, respectively; d represents the feature dimension number. This represents a very small positive number that prevents the denominator from being zero; D is the total dimension of the eigenvectors.

4. The method for aggregating the regulation capability of electric vehicle clusters based on dynamic off-site power boundaries according to claim 3, characterized in that, The electric vehicle clusters described above are divided using the K-means clustering method based on standardized charging behavior feature vectors. The clustering objective function is: ; In the formula: C represents the set of electric vehicle cluster numbers; c represents the electric vehicle cluster number; Representing a scene The set of electric vehicles belonging to cluster c; Let i represent the standardized charging behavior feature vector of the i-th electric vehicle; This represents the cluster center of cluster c; The second norm of a vector; By solving the clustering objective function, electric vehicles with similar charging behavior characteristics are grouped into the same cluster, and the electric vehicle set corresponding to each electric vehicle cluster is obtained.

5. The method for aggregating the regulation capability of electric vehicle clusters based on dynamic off-site power boundaries according to claim 1, characterized in that, The specific calculations for the expected off-site power and the minimum guaranteed off-site power are as follows: ; ; In the formula: Representing a scene The expected off-station charge of the i-th electric vehicle; Representing a scene The minimum guaranteed off-station battery power for the i-th electric vehicle; Indicates the desired state of charge when leaving the station; Indicates the minimum guaranteed off-site charged state; This represents the battery capacity of the i-th electric vehicle; Based on the adjustable margin between the expected off-site charge and the minimum guaranteed off-site charge, determine the maximum acceptable undercharge amount authorized by the user in advance. : 。 6. The method for aggregating the regulation capability of electric vehicle clusters based on dynamic off-site power boundaries according to claim 1, characterized in that, The aforementioned dynamic departure power boundary for a single vehicle includes a lower boundary for dynamic departure power and an upper boundary for vehicle departure power. For vehicle i belonging to electric vehicle cluster c, a dynamic off-station charge boundary relaxation coefficient is introduced to construct the lower boundary of the dynamic off-station charge. as follows: ; In the formula: Representing a scene The expected off-station charge of the i-th electric vehicle; This indicates the maximum acceptable undercharge amount authorized by the user in advance; , representing the dynamic off-site power boundary relaxation coefficient of cluster c, when At that time, the vehicle must reach the original expected departure charge level; when At that time, the vehicle can accept the maximum range of power reduction when leaving the station; Based on the battery's maximum allowable state of charge and the user-allowed additional charging capacity, the upper boundary of the vehicle's off-site charge capacity is constructed as follows: ; ; In the formula: Representing a scene The upper boundary of the off-station battery level of the i-th electric vehicle; Indicates the maximum permissible state of charge of the battery; Indicates battery capacity; Indicates the expected off-site power consumption; This indicates that the user is allowed additional charging beyond the desired off-site charge level; This indicates that the user is allowed additional charging space beyond the expected off-site state of charge; The feasible range of vehicle departure power is defined by the lower boundary of the dynamic departure power and the upper boundary of the departure power: ; In the formula: This indicates the time when the i-th electric vehicle leaves the network. The actual power consumption; and These represent the lower boundary and upper boundary of the dynamic off-site power, respectively.

7. The method for aggregating the regulation capability of electric vehicle clusters based on dynamic off-site power boundaries according to claim 1, characterized in that, Step S3.2 specifically includes: For vehicles that leave the network within the current scheduling cycle, the lower boundary of the dynamic off-site electricity consumption is taken as the required electricity consumption for the current cycle; for vehicles that stay across scheduling cycles, the minimum required electricity consumption at the end of the current cycle is calculated by back-calculating the remaining stay time in the next scheduling cycle, resulting in: ; In the formula: This represents the cut-off electricity demand of the i-th electric vehicle during the current scheduling cycle; Indicates the lower boundary of dynamic off-site power; Indicates the minimum permissible state of charge of the battery; Indicates battery capacity; Indicates charging efficiency; Indicates the maximum charging power; Indicates the extended off-network time; K represents the total number of scheduling periods; Indicates the duration of a single scheduling period; Based on the cutoff demand for electricity and the remaining dwell time of the vehicle, calculate the dynamic minimum acceptable electricity level for the vehicle in time period t: ; In the formula: This represents the minimum acceptable energy level of the i-th electric vehicle during time period t; This indicates the cut-off electricity demand within the current scheduling cycle; Indicates charging efficiency; Indicates the vehicle's maximum charging power; This represents the equivalent off-network time within the current scheduling cycle; Indicates the duration of a single scheduling period; Indicates the minimum permissible state of charge of the battery; Indicates battery capacity; The maximum acceptable battery level for a vehicle during time period t is: ; In the formula: This represents the maximum acceptable battery level of the i-th electric vehicle during time period t; Indicates the upper limit of the vehicle's charge level after leaving the station; Therefore, the current battery level of the vehicle responding online should meet the following requirements: ; In the formula: This represents the current battery level of the i-th electric vehicle during time period t; and These represent the dynamic minimum acceptable battery level and the maximum acceptable battery level, respectively. This refers to the set of vehicles that are online and participating in the response.

8. The method for aggregating the regulation capability of electric vehicle clusters based on dynamic off-site power boundaries according to claim 1, characterized in that, Step S3.3 specifically includes: Calculate the vehicle's state of charge demand consistency rate based on its current battery level relative to the dynamically acceptable battery range: ; In the formula: This represents the consistency rate of the state of charge and demand progress of the i-th electric vehicle in time period t; Indicates the vehicle's current battery level; and These represent the vehicle's minimum and maximum acceptable battery level, respectively. This indicates an extremely small positive number that prevents the denominator from being zero; The lower the state of charge demand progress consistency rate, the closer the vehicle is to the dynamic minimum acceptable charge level, and the more urgent its charging demand is; the higher the state of charge demand progress consistency rate, the higher the vehicle's charge margin, and the stronger its ability to provide discharge regulation.

9. The method for aggregating the regulation capability of electric vehicle clusters based on dynamic off-site power boundaries according to claim 1, characterized in that, Step S3.5 specifically includes: Based on the equal-progress charging and discharging time scales of the electric vehicle cluster, the adjustable charging and discharging capabilities of online and participating vehicles within the cluster are aggregated to obtain the charging power boundary and discharging power boundary of the electric vehicle cluster: ; ; In the formula: and Representing the scene respectively The charging power boundary and discharging power boundary of cluster c in time period t; This represents the set of vehicles within the cluster that are online and participating in the response. and These represent the vehicle's minimum and maximum acceptable battery level, respectively. and These represent the vehicle's charging efficiency and discharging efficiency, respectively. and These represent the time scales for equal-progress charging and equal-progress discharging of the cluster, respectively. The actual charging and discharging power of the cluster must not exceed the corresponding power boundary, and simultaneous charging and discharging are not allowed within the same time period: ; ; ; In the formula: and These represent the actual charging power and discharging power of the cluster, respectively. and These are 0-1 variables representing the charging and discharging states of the cluster, respectively. and Representing the scene respectively The charging power boundary and discharging power boundary of cluster c in time period t.

10. The method for aggregating the regulation capability of electric vehicle clusters based on dynamic off-site power boundaries according to claim 1, characterized in that, In S4, the upper and lower envelopes of the cumulative energy of a single vehicle are calculated in detail as follows; The upper envelope of the cumulative net charging energy of the i-th electric vehicle up to time t is: ; In the formula: This represents the upper envelope of the cumulative net charging energy of the i-th electric vehicle up to time period t; Indicates the time of network access; Indicates the equivalent time of disconnection; Indicates battery capacity; Indicates the upper boundary of the off-site power quantity; Indicates the initial state of charge of the vehicle upon grid connection; Indicates vehicle charging efficiency; Indicates the vehicle's maximum charging power; Indicates the duration of a single scheduling period; The lower envelope of the cumulative net charging energy of the i-th electric vehicle up to time t is: ; In the formula: This represents the lower envelope of the cumulative net charging energy of the i-th electric vehicle up to time period t; This represents the dynamic minimum acceptable power consumption for time period t. This indicates the cut-off electricity demand within the current scheduling cycle; Indicates the initial state of charge of the vehicle upon grid connection; Indicates the time of network access; Indicates the equivalent time of disconnection; This indicates the battery capacity.

11. The method for aggregating the regulation capability of electric vehicle clusters based on dynamic off-site power boundaries according to claim 10, characterized in that, The specific calculation of the cluster's cumulative net charging energy in step S4 is as follows: ; In the formula: This represents the cumulative net charging energy of cluster c up to time period t; k represents the scheduling period number in the cumulative calculation. and These represent the charging power and discharging power of the cluster during time period k, respectively. and These represent the cluster's equivalent charging efficiency and equivalent discharging efficiency, respectively. Indicates the scheduling time interval.

12. The method for aggregating the regulation capability of electric vehicle clusters based on dynamic off-site power boundaries according to claim 11, characterized in that, The specific steps for calculating the cluster's cumulative net charging energy in step S4 are as follows; By summing the upper and lower envelopes of the cumulative net charging energy of each vehicle within the same electric vehicle cluster, we obtain the lower and upper boundaries of the cluster's cumulative energy. ; ; In the formula: and Representing the scene respectively The lower boundary and upper boundary of the cumulative energy of cluster c in time period t; This represents the set of electric vehicles participating in the response within cluster c; and These represent the lower envelope of the cumulative net charging energy and the upper envelope of the cumulative net charging energy for vehicle i up to time period t, respectively. The cumulative net charging energy of the electric vehicle cluster meets: ; In the formula: Representing a scene The cumulative net charging energy of cluster c up to time period t.

13. The method for aggregating the regulation capability of electric vehicle clusters based on dynamic off-site power boundaries according to claim 12, characterized in that, The specific calculation results of the electric vehicle cluster regulation capability aggregation in S4 are as follows; The cluster charging power boundary, discharging power boundary, lower boundary of cumulative energy, and upper boundary of cumulative energy for each scheduling period are combined to form the electric vehicle cluster regulation capability boundary: ; In the formula: Representing a scene The set of adjustment capability boundaries for cluster c; and These represent the cluster charging power boundary and discharging power boundary for time period t, respectively. and These represent the lower boundary and upper boundary of the cluster's cumulative energy up to time period t, respectively. This represents the set of scheduling periods.

14. A system for aggregating the regulation capacity of an electric vehicle cluster based on a dynamic off-site power boundary, characterized in that, include: The electric vehicle data processing and cluster partitioning module is used to obtain electric vehicle operating parameters, construct and standardize charging behavior feature vectors, and partition electric vehicle clusters. The single-vehicle dynamic departure power boundary construction module is used to construct the single-vehicle dynamic departure power boundary based on the expected departure power, the minimum guaranteed departure power, and the maximum acceptable undercharge amount authorized by the user. The vehicle cluster charging and discharging power boundary aggregation module is used to aggregate the vehicle cluster charging power boundary and discharging power boundary based on the vehicle online response status, dynamic acceptable power range and the consistency rate of state of charge demand progress. The vehicle cluster cumulative energy boundary aggregation module is used to aggregate the upper and lower envelopes of the cumulative net charging energy of a single vehicle, obtain the upper and lower boundaries of the cumulative energy of the vehicle cluster, and form the aggregation result of the electric vehicle cluster regulation capability.

15. An electronic device comprising a processor and a memory for storing computer program instructions, characterized in that: The processor executes this instruction to implement the electric vehicle cluster regulation capability aggregation method based on dynamic off-site power boundary as described in any one of claims 1-13.

16. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed, it can realize the electric vehicle cluster regulation capability aggregation method based on dynamic off-site power boundary as described in any one of claims 1-13.

Citation Information

Patent Citations

  • Determination method, scheduling method and scheduling system for regulatable capability of electric vehicle cluster

    CN113541168A

  • Electric vehicle cluster scheduling method and device, storage medium and electronic device

    CN117592700A