Multi-zone-area electric vehicle cluster scheduling optimization method and device and computer equipment

Abstract

The invention provides a multi-zone-area electric vehicle cluster scheduling optimization method and device and computer equipment, and the method comprises the steps: determining the individual type of each electric vehicle in a multi-zone-area electric vehicle cluster according to the vehicle parameters of each electric vehicle, and enabling the individual types to comprise a non-scheduling type, a transferable type and a chargeable and dischargeable type; based on the vehicle parameters and the individual type of each electric vehicle, establishing an individual feasible region of each electric vehicle, and according to each individual feasible region, obtaining a cluster feasible region of the electric vehicle cluster in the court by adopting the Mono polyhedron modeling and Minkowski summation; based on robust optimization constraint conditions, multi-zone-area electric vehicle cluster scheduling is carried out in a cluster feasible region, and the robust optimization constraint conditions are used for ensuring that zone-area transformers are not overloaded when electric vehicle charging behaviors are scheduled, and meanwhile, safety redundancy of family load extreme changes is reserved. According to the method, individual type division and cluster feasible region modeling are introduced, so that the problem of uncertainty in the scheduling process is effectively relieved.

Description

Technical Field

[0001] The present application relates to the technical field of power system optimization and dispatching, and in particular to a multi-station electric vehicle cluster dispatching optimization method, device and computer equipment. Background Art

[0002] The rapid growth of electric vehicles (EVs), particularly with the widespread integration of EV clusters across multiple substations, is increasing operational pressure on the power grid. The highly random and time-varying nature of EV loads, combined with the volatility of other residential and commercial loads within the substation, forces grid dispatchers to address both load-side and supply-side uncertainties.

[0003] Currently, research on the uncertainty problem in EV cluster scheduling mainly focuses on two mainstream methods. One is stochastic optimization, which is based on known or estimable probability distributions and makes optimization decisions by minimizing expected costs or maximizing expected benefits. For example, this method predicts demand distribution based on historical charging behavior and formulates charging strategies. The other is robust optimization, which does not rely on probability distributions, but seeks scheduling solutions that are still feasible in the worst case scenario within the uncertain set, such as constructing new energy fluctuation ranges to improve scheduling safety in extreme weather.

[0004] While these approaches have mitigated the uncertainty issues in scheduling to some extent, they still face numerous challenges in practical application. From the perspective of uncertainty modeling, stochastic optimization relies heavily on probability distributions, making it difficult to cope with situations where data is insufficient or distributions vary. While robust optimization enhances security in extreme scenarios, most methods focus on uncertainty in only a single dimension, failing to fully capture the spatiotemporal coupling of factors such as user demand and renewable energy output. This makes it difficult for scheduling strategies to maintain global adaptability and stability under multiple uncertainties. Summary of the Invention

[0005] The purpose of this application is to solve at least one of the above-mentioned technical deficiencies, especially the technical deficiency in the prior art that scheduling strategies are difficult to maintain global adaptability and stability under multiple uncertain conditions.

[0006] In a first aspect, the present application provides a multi-area electric vehicle cluster scheduling optimization method, the method comprising:

[0007] According to the vehicle parameters of each electric vehicle, the individual type of each electric vehicle in the multi-area electric vehicle cluster is determined, and the individual types include non-dispatchable type, transferable type and rechargeable and dischargeable type;

[0008] Based on the vehicle parameters and individual type of each electric vehicle, the individual feasible region of each electric vehicle is established. Based on each individual feasible region, the cluster feasible region of the electric vehicle cluster in the substation is obtained by using Chino polyhedron modeling and Minkowski summation.

[0009] Based on preset robust optimization constraints, multi-station electric vehicle cluster scheduling is carried out in the cluster feasible domain. The robust optimization constraints are used to ensure that the station transformers are not overloaded when scheduling electric vehicle charging behavior, while retaining safety redundancy for extreme changes in household loads.

[0010] In one embodiment, the vehicle parameters include a current power value, an expected power value, a rated charging power, a charging power, and a discharging power; and the step of determining the individual type of each electric vehicle in the multi-area electric vehicle cluster according to the vehicle parameters of each electric vehicle comprises:

[0011] For each electric vehicle, if the current power value of the electric vehicle is not less than the corresponding expected power value, and the charging power of the electric vehicle is equal to the corresponding rated charging power, then the individual type of the electric vehicle is non-dispatchable;

[0012] If the charging power of the electric vehicle is an adjustable charging power, determining whether the electric vehicle is prohibited from discharging;

[0013] When the electric vehicle is prohibited from discharging, the individual type of the electric vehicle is a transferable type, and when the electric vehicle is allowed to discharge, the individual type of the electric vehicle is a chargeable and dischargeable type.

[0014] In one embodiment, the unschedulable expression is:

[0015]

[0016] The transferable expression is:

[0017]

[0018] The expression for the rechargeable and dischargeable type is:

[0019]

[0020] in, represents the charging power of the non-dispatchable electric vehicle at time t, Indicates the rated charging power of the electric vehicle, represents the power value of the electric vehicle at time t in the non-dispatchable type, Indicates the expected power value, Indicates the maximum power value. represents the power value of the electric vehicle at time t0 in the non-dispatchable type, Indicates the charging efficiency, Indicates the charging time interval, represents the charging power of the transferable electric vehicle at time t, Indicates fast charging power. Indicates slow charging power. represents the power value of the transferable electric vehicle at time t, Indicates the minimum power value. represents the power value of the transferable electric vehicle at time t0, represents the charging power of the rechargeable and dischargeable electric vehicle at time t, Indicates the discharge power, Indicates the power value of a rechargeable and dischargeable electric vehicle at time t, Indicates the power value of a rechargeable and dischargeable electric vehicle at time t0, represents the discharge efficiency, represents the dispatchable power, Indicates the discharge capacity of a rechargeable electric vehicle.

[0021] In one embodiment, the vehicle parameters include a current power value and a minimum power value; and the step of establishing an individual feasible domain for each electric vehicle based on the vehicle parameters and individual type of each electric vehicle includes:

[0022] For electric vehicles with non-dispatchable individual types, the individual feasible region of the electric vehicle is non-dispatchable capacity and non-dispatchable power;

[0023] For an electric vehicle of transferable individual type, when the current power value of the electric vehicle does not reach the corresponding minimum power value, the individual feasible domain of the electric vehicle is non-dispatchable capacity and non-dispatchable power. When the current power value of the electric vehicle reaches the corresponding minimum power value, the individual feasible domain of the electric vehicle is only dispatchable power and the dispatchable capacity is zero. The dispatchable power in the individual feasible domain of the electric vehicle is controlled by a preset first satisfaction parameter.

[0024] For an electric vehicle whose individual type is rechargeable and dischargeable, when the current power value of the electric vehicle does not reach the corresponding minimum power value, the individual feasible domain of the electric vehicle is non-dispatchable capacity and non-dispatchable power; when the current power value of the electric vehicle reaches the corresponding minimum power value, the individual feasible domain of the electric vehicle is dispatchable power and dispatchable capacity, wherein the dispatchable power and dispatchable capacity in the individual feasible domain of the electric vehicle are controlled by a preset second satisfaction parameter.

[0025] In one embodiment, the steps of obtaining the cluster feasible region of the electric vehicle cluster in the substation area by using Chino polyhedron modeling and Minkowski summation according to each individual feasible region include:

[0026] A Qino polyhedron model of each individual feasible region is constructed, and Minkowski summation is performed on each Qino polyhedron model to obtain the cluster feasible region.

[0027] In one embodiment, the expression of the Qino polyhedron model of each individual feasible region is:

[0028]

[0029] in, represents the center point vector of the i-th Qino polyhedron model, Indicates the end time. Indicates the initial time, Indicates the maximum power value. Indicates the minimum power value. represents the first generating vector of the i-th Qino polyhedron model, Indicates fast charging power. Represents the second generating vector of the i-th Qino polyhedron model.

[0030] In one embodiment, the expression of the robust optimization constraint is:

[0031]

[0032] in, represents the charging power of the i-th electric vehicle at time t, Indicates the safety margin of the station area, Indicates the total number of electric vehicles in the area. Indicates the rated power of the transformer in the substation area. represents the prediction error, Represents the predicted household load power.

[0033] In a second aspect, the present application provides a multi-area electric vehicle cluster scheduling optimization device, the device comprising:

[0034] An individual type determination module is used to determine the individual type of each electric vehicle in the multi-area electric vehicle cluster according to the vehicle parameters of each electric vehicle, and the individual types include non-dispatchable type, transferable type and rechargeable and dischargeable type;

[0035] The cluster feasible region determination module is used to establish the individual feasible region of each electric vehicle based on the vehicle parameters and individual type of each electric vehicle, and to obtain the cluster feasible region of the electric vehicle cluster in the substation by using Chino polyhedron modeling and Minkowski summation based on each individual feasible region;

[0036] The electric vehicle cluster scheduling module is used to schedule electric vehicle clusters in multiple substations within the cluster feasible domain based on preset robust optimization constraints. The robust optimization constraints are used to ensure that the substation transformers are not overloaded when scheduling electric vehicle charging behavior, while retaining safety redundancy for extreme changes in household loads.

[0037] In a third aspect, the present application provides a storage medium: the storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the multi-station electric vehicle cluster scheduling optimization method as described in any one of the above embodiments.

[0038] In a fourth aspect, the present application provides a computer device, comprising: one or more processors, and a memory;

[0039] The memory stores computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the steps of the multi-area electric vehicle cluster scheduling optimization method in any one of the above embodiments are executed.

[0040] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0041] The multi-station electric vehicle cluster scheduling optimization method provided in this application effectively alleviates the uncertainty problem in the scheduling process by introducing individual type classification and cluster feasible domain modeling. First, by dividing electric vehicles into three categories: non-dispatchable, transferable, and rechargeable, the flexibility of various types of electric vehicles in scheduling is fully explored, which is conducive to the refined management of resources; secondly, based on individual parameters, a feasible domain is established, and the cluster feasible domain is constructed using Chino polyhedron modeling and Minkowski summation, which can systematically integrate the overall regulation capability of the electric vehicle cluster and accurately reflect the coupling characteristics and flexible boundaries of the group in time and space; finally, robust optimization constraints are introduced in the scheduling decision, which does not rely on a specific probability distribution, enhances the adaptability and stability of the scheduling strategy in the face of residential load fluctuations and extreme situations, and ensures that the transformer in the station is not overloaded while retaining the necessary safety redundancy. Overall, this method achieves an effective balance between supply security, scheduling flexibility, and uncertainty resistance, and significantly improves the practicality and reliability of the multi-station electric vehicle cluster scheduling system. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0043] Figure 1 A flow chart of a multi-zone electric vehicle cluster scheduling optimization method provided in an embodiment of the present application;

[0044] Figure 2 A schematic diagram of the structure of a multi-zone electric vehicle cluster scheduling optimization device provided in an embodiment of the present application;

[0045] Figure 3 A schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0047] This application provides a multi-station electric vehicle cluster scheduling optimization method. The following embodiments are described using the method applied to a computer device as an example. It is understood that the computer device can be any device with data processing capabilities, including but not limited to a single server, a server cluster, a personal laptop computer, a desktop computer, etc. Figure 1 As shown, the method may include the following steps:

[0048] S101: Determine the individual type of each electric vehicle in a multi-area electric vehicle cluster based on vehicle parameters of each electric vehicle, where the individual types include a non-dispatchable type, a transferable type, and a chargeable and dischargeable type.

[0049] Among them, vehicle parameters refer to various types of information used to describe the status and operating characteristics of electric vehicles, including but not limited to battery capacity, current power, charging status, user-defined charging time window, vehicle connection duration, the vehicle's current geographical area, historical charging behavior data, etc. Individual type refers to the classification of electric vehicles based on their ability to respond to scheduling strategies within a given scheduling cycle, specifically including non-schedulable, transferable, and rechargeable. The non-schedulable type means that no scheduling instructions will be accepted within the current cycle, and only independent charging will be performed according to user settings. The transferable type means that charging tasks can be flexibly transferred within the user-defined time window, but do not have the ability to discharge. The rechargeable and dischargeable type means that not only can the charging time be flexibly transferred, but also the ability to discharge to the power grid.

[0050] First, by communicating with electric vehicle charging stations, energy management systems, or vehicle terminal devices, the system periodically or in real time acquires the vehicle parameter data of each electric vehicle connected to the power grid in the substation area. This data is then cached in a local database or reported to a central dispatch server. During this process, standardized protocols such as OCPP or customized IoT protocols can be used to achieve data exchange and ensure a consistent data format across different brands and models of vehicles.

[0051] After obtaining the vehicle parameters, the computer device determines the dispatchability of each electric vehicle using a preset classification rule model. This model can be implemented based on logical judgment rules or machine learning classifiers. For example, if the expected connection time is less than the minimum dispatchable threshold or the user has not authorized participation in the dispatch, the vehicle is judged as non-dispatchable; if the current battery level is less than the minimum discharge threshold and the remaining connection time is greater than the minimum charging time, the vehicle is judged as transferable; if the current battery level is greater than the dischargeable threshold and the user has authorized participation in the V2G service, the vehicle is judged as chargeable and dischargeable. Based on this judgment logic, each vehicle is labeled and a structured dispatch information set is formed.

[0052] Once individual types are labeled, the classification results are bound to the corresponding vehicle's unique identification code and stored in a database or cache system for subsequent scheduling optimization. This labeling is not only used to construct the cluster's feasible domain, but also to generate refined scheduling plans. For example, prioritizing transferable vehicles to share peak loads or deploying rechargeable and dischargeable vehicles to support load regulation when power grid supply and demand are unbalanced. This structured individual capability information enables the scheduling system to have good controllability and responsiveness when dealing with the heterogeneous characteristics of electric vehicle clusters.

[0053] Determining individual types based on vehicle parameters enables accurate identification of the dispatchability of electric vehicles, making dispatch strategies more targeted in resource allocation and avoiding the potential issues of unresponsiveness, conflict, or load imbalance that may arise from a one-size-fits-all dispatch of all vehicles. It also enables a coordinated match between vehicle behavior characteristics and the operating status of the substation. Under the premise that the dispatch response capabilities of different types of vehicles are accurately calibrated, a hierarchical strategy can be developed to improve response efficiency, reduce dispatch errors, and provide a solid data foundation for the subsequent construction of cluster feasible domains and robust dispatch. Therefore, this classification process not only improves the intelligence level of electric vehicle dispatch, but also provides stronger stability and security guarantees in an environment with fluctuating and uncertain supply and demand in the power grid.

[0054] S102: Based on the vehicle parameters and individual type of each electric vehicle, establish the individual feasible region of each electric vehicle, and use Chino polyhedron modeling and Minkowski summation according to each individual feasible region to obtain the cluster feasible region of the electric vehicle cluster in the substation.

[0055] Among them, the individual feasible domain refers to the set of feasible operations for each electric vehicle, constructed according to its vehicle parameters and its corresponding individual type, for the change of the vehicle's charging and discharging power and energy within a certain time range, which can be expressed as a high-dimensional constraint polyhedron. Chino polyhedron is a mathematical modeling method for expressing non-convex geometric sets, which is used to describe the scheduling feasible domain under multiple discontinuous constraint combinations. Minkowski summation refers to merging multiple individual feasible domains into an overall feasible domain through vector space summation, which is used to construct the cluster feasible domain, that is, the joint scheduling operation space that multiple electric vehicles can form in a certain area. It is the basic constraint area for group scheduling optimization.

[0056] The computer device accesses the individual data streams of all electric vehicles in the area. Through a unified interface module, it obtains real-time or predictive vehicle parameters for each electric vehicle, including but not limited to SOC, remaining connection time, battery capacity, and historical charging curves. Combined with previously determined individual type labels, this creates the original input dataset for modeling.

[0057] The computer device uses a built-in constraint generation module to construct an individual feasible domain for each vehicle. For example, for a transferable vehicle, its feasible domain can be constructed from multiple linear inequality constraints, forming a convex set, based on parameters such as the time window, minimum / maximum charging power, and desired power. For a rechargeable and dischargeable vehicle, its feasible domain includes not only charging constraints but also discharging constraints and power reverse conversion conditions. The resulting feasible domain may be non-convex. Therefore, a Chino polyhedron modeling method is used to express these various conditions and multi-state regions as a set representation of piecewise linear polyhedra, improving the accuracy and flexibility of the feasible domain representation. The computer device uses a constraint construction algorithm to output the corresponding high-dimensional geometric feasible set for each vehicle.

[0058] After constructing the individual feasible regions for each vehicle, the computer invokes the summation engine module to perform a Minkowski summation on all individual feasible regions, thereby synthesizing the cluster feasible region for the entire station area. During this process, parallel computing strategies are employed to calculate the summation results for different vehicle groups. Constraint merging and dimensionality reduction strategies are then used to optimize the summation results, ensuring mathematical closure and scheduling practicality. The resulting cluster feasible region serves as the joint scheduling boundary space, which can be subsequently applied to the robust scheduling model.

[0059] By constructing an individual feasible region for each electric vehicle based on vehicle parameters and individual type, and further employing Chino polyhedron modeling and Minkowski summation to generate a cluster feasible region for scheduling modeling, this approach effectively addresses the diverse and complex nature of electric vehicle groups in terms of behavior patterns and scheduling capabilities. This approach not only improves the precision and flexibility of expressing individual vehicle constraints but also ensures the mathematical feasibility and engineering feasibility of group scheduling solutions through structured synthesis of overall scheduling boundaries. Particularly in high-dimensional and complex scheduling scenarios, this approach avoids the dimensionality explosion associated with directly mixing all individual vehicles for optimization. It also enables the scheduling model to fully absorb and utilize the diverse characteristics of electric vehicles, enhancing its robust adaptability to load fluctuations and dynamic resource changes. Therefore, while improving the efficiency of scheduling optimization solutions, it also ensures the global feasibility and safety of scheduling strategies in uncertain environments.

[0060] S103: Based on the preset robust optimization constraints, multi-station electric vehicle cluster scheduling is performed in the cluster feasible domain. The robust optimization constraints are used to ensure that the station transformer is not overloaded when scheduling electric vehicle charging behavior, while retaining safety redundancy for extreme changes in household loads.

[0061] Robust optimization constraints refer to a set of pre-defined constraints with fault-tolerant boundaries that are used by computers to manage external disturbances such as household load uncertainty, grid load fluctuations, and measurement errors when scheduling charging groups of electric vehicles. These constraints typically include the maximum allowable load threshold for transformers in a substation, the range of reserved safety redundant capacity, and upper and lower bounds for peak and valley load fluctuations. These constraints ensure that the scheduling plan does not violate hard power supply constraints even under the most unfavorable circumstances. Multi-substation electric vehicle cluster scheduling refers to the joint charging control decision-making process for electric vehicles within multiple distribution substations. Its goal is to optimize the overall scheduling strategy while satisfying local distribution capacity constraints and global grid load coordination objectives.

[0062] The computer receives cluster feasible region data from each substation and simultaneously loads the robust optimization constraints corresponding to each substation's transformers. The dispatch control module establishes a constrained optimization model for the entire EV fleet's scheduling problem. The model's objective function can be set to minimize electricity price costs or peak load. The feasible space is defined by the feasible region of each cluster, and transformer capacity limitations and load uncertainty buffers are introduced as robust constraints. This ensures that the final dispatch solution is not only feasible under ideal conditions but also executable under load disturbances.

[0063] The computer equipment further calls the internal robust optimization solution module to explicitly convert the robust constraints into worst-case scheduling conditions on the uncertainty set. This module sets an adjustment boundary with a safety margin for the load upper limit of each substation transformer, such as setting a redundancy threshold at 90% of the base maximum load. At the same time, it analyzes the historical fluctuation range of household load to construct an uncertain disturbance set. By applying this type of constraint to the power scheduling variables of each time slice, a multi-stage optimization problem with uncertain parameters is constructed and solved in the cluster feasible domain using algorithms such as branch and bound, scenario approximation, or variable structure interior point method to ensure a certain degree of robustness.

[0064] Once the solution is complete, the computer equipment will distribute the dispatch results to the electric vehicle terminal controllers in each substation, retaining a dynamic feedback module for real-time corrections. While the electric vehicles are executing the dispatch command, if abnormal household loads are detected, such as a sudden increase in a user's load, robust optimization is immediately used to reserve redundancy for peak load avoidance. If necessary, the optimization model can be re-solved to adapt to the new actual load information, achieving rolling dispatch control. This process ensures that the group regulation potential of electric vehicles is maximized while ensuring power supply security.

[0065] By scheduling within the cluster's feasible region and introducing robust optimization constraints to limit scheduling behavior, computer equipment can fully utilize the regulation capabilities of the electric vehicle group while effectively preventing overloads on substation transformers caused by centralized charging. Furthermore, by retaining redundancy for extreme variations in household loads, the scheduling strategy is able to cope with sudden load disturbances, improving reliability and resilience. This approach ensures that scheduling solutions remain effective and feasible in the face of unpredictable real-world situations, reduces the risk of power outages caused by overload protection triggering, improves the user charging experience, and promotes smoothing of grid loads. Therefore, this execution method achieves a good balance between ensuring grid security and ensuring stable energy supply to users, and has significant engineering practical value and economic benefits.

[0066] In the above embodiment, by introducing individual type classification and cluster feasible domain modeling, the uncertainty problem in the scheduling process is effectively alleviated. First, by dividing electric vehicles into three categories: non-dispatchable, transferable, and rechargeable, the flexibility of various types of electric vehicles in scheduling is fully explored, which is conducive to the realization of refined resource management; secondly, based on the establishment of a feasible domain of individual parameters, and using Chino polyhedron modeling and Minkowski summation to construct the feasible domain of the cluster, the overall regulation capability of the electric vehicle cluster can be systematically integrated, and the coupling characteristics and flexible boundaries of the group in time and space can be accurately reflected; finally, the robust optimization constraint is introduced in the scheduling decision, which does not rely on a specific probability distribution, and enhances the adaptability and stability of the scheduling strategy in the face of residential load fluctuations and extreme situations, ensuring that the transformer in the substation is not overloaded while retaining the necessary safety redundancy. Overall, this method achieves an effective balance between supply security, scheduling flexibility, and uncertainty resistance, and significantly improves the practicality and reliability of the multi-substation electric vehicle cluster scheduling system.

[0067] In one embodiment, the vehicle parameters include a current charge value, an expected charge value, a rated charging power, a charging power, and a discharging power; and the step of determining the individual type of each electric vehicle in the multi-area electric vehicle cluster according to the vehicle parameters of each electric vehicle includes:

[0068] For each electric vehicle, if the current power value of the electric vehicle is not less than the corresponding expected power value, and the charging power of the electric vehicle is equal to the corresponding rated charging power, then the individual type of the electric vehicle is non-dispatchable;

[0069] If the charging power of the electric vehicle is an adjustable charging power, determining whether the electric vehicle is prohibited from discharging;

[0070] When the electric vehicle is prohibited from discharging, the individual type of the electric vehicle is a transferable type, and when the electric vehicle is allowed to discharge, the individual type of the electric vehicle is a chargeable and dischargeable type.

[0071] Among them, the current power value refers to the current power value stored in the vehicle's power battery as collected by the electric vehicle terminal, measured in kilowatt-hours (kWh). The expected power value refers to the minimum power threshold preset by the vehicle user or system based on factors such as the next travel demand and planned driving distance. The rated charging power refers to the standard maximum safe charging power specified by the vehicle manufacturer. The adjustable charging power means that the charging power currently supported by the vehicle is below the rated value and can be adjusted up or down. Discharge prohibition means that the control strategy set by the system or user explicitly restricts the vehicle from outputting electrical energy to the grid.

[0072] The vehicle network interface periodically receives status parameters uploaded by each electric vehicle, including the current power level, the user-set desired power level, the current charging power, whether the discharge function is enabled, whether the vehicle is connected to a charging station, and other information. After receiving this data, the power parameters are first judged: if the vehicle's current power level is greater than or equal to the desired power level and the charging power has reached the rated value, it indicates that the vehicle cannot provide further adjustment capabilities. The individual type is determined to be "unschedulable" and marked as a controllable object that is not included in this round of scheduling optimization model.

[0073] For vehicles not identified as non-dispatchable, further analysis is performed to determine whether their current charging power is adjustable. If the charging power is less than the rated value, this indicates that there is still room for power adjustment, and the vehicle is classified as adjustable. Next, the vehicle's discharge permission parameters are read. If discharge is prohibited, the vehicle can only shift charging power to fill peaks and valleys, and is classified as transferable. Conversely, if discharge is permitted, the vehicle can participate in V2G (Vehicle to Grid) interactions, both charging and unloading energy, and is classified as "chargeable and dischargeable."

[0074] After individual vehicle types are determined, each type of vehicle is stored in different dispatch participation queues according to its type. Non-dispatched vehicles are excluded from dispatch; transferable vehicles participate in peak-valley regulation strategies for optimizing electricity prices or shifting loads; and rechargeable vehicles are included in the two-way energy trading dispatch unit, providing discharge support or temporary energy storage when needed. This type identification not only provides accurate input for subsequent grid load optimization and dispatch, but also ensures the appropriate use of vehicles of varying capabilities, avoiding energy losses or safety risks caused by excessive control.

[0075] By determining whether the current power level has reached the expected level and whether the current charging power has reached the rated power, it is possible to quickly identify non-dispatchable electric vehicles that lack adjustment space, thereby reducing the scheduling calculation burden. Furthermore, based on the determination of power adjustment capability and discharge authority, the remaining vehicles can be divided into transferable and rechargeable individuals with control value, which is conducive to building a refined and hierarchical scheduling system. This classification strategy improves scheduling accuracy and resource utilization efficiency, making group scheduling control more dynamic and responsive, while taking into account user needs and grid security, and helping to achieve orderly access to distributed power sources and flexible management of local loads.

[0076] In one embodiment, the unschedulable expression is:

[0077]

[0078] The transferable expression is:

[0079]

[0080] The expression for the rechargeable and dischargeable type is:

[0081]

[0082] in, represents the charging power of the non-dispatchable electric vehicle at time t, Indicates the rated charging power of the electric vehicle, represents the power value of the electric vehicle at time t in the non-dispatchable type, Indicates the expected power value, Indicates the maximum power value. represents the power value of the electric vehicle at time t0 in the non-dispatchable type, Indicates the charging efficiency, Indicates the charging time interval, represents the charging power of the transferable electric vehicle at time t, Indicates fast charging power. Indicates slow charging power. represents the power value of the transferable electric vehicle at time t, Indicates the minimum power value. represents the power value of the transferable electric vehicle at time t0, represents the charging power of the rechargeable and dischargeable electric vehicle at time t, Indicates the discharge power, Indicates the power value of a rechargeable and dischargeable electric vehicle at time t, Indicates the power value of a rechargeable and dischargeable electric vehicle at time t0, represents the discharge efficiency, represents the dispatchable power, Indicates the discharge capacity of a rechargeable electric vehicle.

[0083] Specifically, for the unschedulable type, , indicating that the electric vehicle is currently charging at the rated charging power, indicating that its power cannot be adjusted; , indicating that the current power level has met user expectations and has not exceeded the maximum battery capacity; , describing the dynamic evolution of charge as charging power changes. The charging efficiency η1 affects the actual charge growth. This type of electric vehicle already has enough charge to meet basic vehicle usage needs, and its charging power is fixed and unadjustable, meaning it cannot play a regulatory role in the charge-discharge scheduling strategy, thus being classified as non-schedulable.

[0084] For transferable types, It can switch between fast charging, slow charging or stopping, indicating that the power has the ability to adjust; when Force fast charging to prevent excessive power loss; Operate within the adjustable range and maintain a reasonable power window; power changes according to charging efficiency Update. These electric vehicles cannot discharge and can only participate in grid dispatch by adjusting their charging rate, such as load shifting or peak-shaving, and are therefore defined as transferable.

[0085] For rechargeable and dischargeable types, Including charging, discharging, and intermediate states (0), with bidirectional regulation capability; the power is within the allowable regulation range; different efficiencies are used according to the charging and discharging direction or Update SOC; This represents the currently available discharge capacity, i.e., the portion of the charge above the minimum charge. This type of electric vehicle is the most flexible to regulate, as it can not only adjust the charging power but also discharge to support the power grid, and is therefore defined as rechargeable.

[0086] It can be understood that the formal modeling approach clearly describes the charging and discharging capabilities and operating constraints of different types of electric vehicles, providing a theoretical basis and implementation path for the classified scheduling of electric vehicles. They can accurately identify whether a vehicle has the ability to adjust charging power or discharge power, thereby realizing hierarchical control of non-dispatchable, transferable, and rechargeable and dischargeable vehicles. By modeling power changes as dynamic processes related to power, efficiency, and time intervals, the expression provides a predictable state evolution basis for the scheduling strategy, ensuring that the overall energy management is optimized while meeting user power requirements. This classified modeling approach not only improves the flexibility and response speed of electric vehicle aggregation control, but also significantly enhances its value and efficiency in participating in grid scheduling as a distributed adjustable resource.

[0087] The three types of formulas in this embodiment accurately distinguish the charging behavior of electric vehicles from the dispatching behavior of the power grid, so that classified management and hierarchical control can be achieved according to the vehicle status, avoiding resource waste or equipment loss caused by blind unified control. In the dispatching process, by determining whether it has power regulation capability or discharge capability, it helps to screen out the target objects that are most valuable for dispatching, thereby improving the flexibility, stability and energy efficiency of the overall dispatching. At the same time, combined with the dynamic modeling of charging and discharging efficiency, it can truly reflect the loss in the energy transmission process and improve the accuracy of dispatching decisions and the safety of execution. Therefore, the execution of this expression not only helps to reduce the complexity of system operation, but also improves the efficiency and reliability of electric vehicles participating in the operation of the power system as distributed adjustable resources.

[0088] In one embodiment, the vehicle parameters include a current power value and a minimum power value; and the step of establishing an individual feasible domain for each electric vehicle based on the vehicle parameters and individual type of each electric vehicle includes:

[0089] For electric vehicles with non-dispatchable individual types, the individual feasible region of the electric vehicle is non-dispatchable capacity and non-dispatchable power;

[0090] For an electric vehicle of transferable individual type, when the current power value of the electric vehicle does not reach the corresponding minimum power value, the individual feasible domain of the electric vehicle is non-dispatchable capacity and non-dispatchable power. When the current power value of the electric vehicle reaches the corresponding minimum power value, the individual feasible domain of the electric vehicle is only dispatchable power and the dispatchable capacity is zero. The dispatchable power in the individual feasible domain of the electric vehicle is controlled by a preset first satisfaction parameter.

[0091] For an electric vehicle whose individual type is rechargeable and dischargeable, when the current power value of the electric vehicle does not reach the corresponding minimum power value, the individual feasible domain of the electric vehicle is non-dispatchable capacity and non-dispatchable power; when the current power value of the electric vehicle reaches the corresponding minimum power value, the individual feasible domain of the electric vehicle is dispatchable power and dispatchable capacity, wherein the dispatchable power and dispatchable capacity in the individual feasible domain of the electric vehicle are controlled by a preset second satisfaction parameter.

[0092] Among them, the individual feasible domain refers to the range of charging and discharging capabilities that electric vehicles can provide under the current state and strategy constraints, specifically including the two dimensions of dispatchable power and dispatchable capacity; non-dispatchable power and non-dispatchable capacity represent the parts that are currently unavailable for dispatch control; the minimum power value is the lower limit of power to ensure the basic travel needs of the vehicle; the satisfaction parameter is an indicator used to measure the owner's or system's acceptance of the flexibility of vehicle participation in dispatching. The first satisfaction parameter is used to control the power dispatch range of transferable vehicles, and the second satisfaction parameter is used to simultaneously regulate the power and capacity dispatch range of rechargeable and dischargeable vehicles.

[0093] Once the individual type is identified, the corresponding individual feasible domain is calculated based on the vehicle's current power level and dispatch capability. For non-dispatchable vehicles, since they meet the power expectations and their power cannot be adjusted, their individual feasible domain will be marked as non-dispatchable power and non-dispatchable capacity, and they will not be included in the dispatch resource set in subsequent dispatch.

[0094] For transferable vehicles, if the current power value does not reach the minimum power value, it means that the vehicle has not yet met basic travel needs, so its participation in scheduling will be regarded as an unsafe operation and will also be marked as unschedulable power and capacity; when the power value reaches the minimum power value, the vehicle will be transferred to the scheduling participation state, and its dispatchable power will be adjusted through the preset first satisfaction parameter to control the upper limit of the power that the vehicle can provide, ensuring a balance between its charging time and user expectations.

[0095] For rechargeable and dischargeable vehicles, when the current power level has not reached the minimum power value, its status is similar to that of the transferable type, and scheduling participation is not allowed; when it reaches the minimum power value, its dispatchable power and dispatchable capacity are calculated respectively according to the second satisfaction parameter, and it is regulated to participate in charging and discharge when the grid load requires, maximizing the value of electric vehicles as adjustable bidirectional energy storage units.

[0096] This process can be executed cyclically by computer equipment within a predetermined scheduling cycle. Through state collection, type judgment, feasible domain calculation and satisfaction constraint processes, a dynamic scheduling resource pool is formed to support dynamic switching and participation degree control of different vehicle states.

[0097] In one example, in order to achieve coordinated dispatch of electric vehicle (EV) clusters at the substation level, it is necessary to build the overall substation feasible region based on the dispatchable capacity and dispatchable power models of the three types of EVs and the fluctuating demand of household loads in the substation, and calculate the dispatchable capacity and dispatchable power of the substation at the current moment. In this process, non-dispatchable EVs and EVs that have not yet reached the minimum expected power are Transferable and rechargeable EVs are regarded as non-dispatchable resources, and their corresponding capacity and power are not available for dispatch.

[0098] When the transferable EV reaches its When , its status is considered to be able to participate in scheduling, but only as a dispatchable power resource, and its dispatchable capacity is zero. On this basis, according to the power constraint conditions, its dispatchable power The following relationship is satisfied:

[0099]

[0100] In order to better balance the contradiction between user participation in scheduling and personal charging needs, the satisfaction parameter is introduced. and control factors :

[0101]

[0102] when When it is 0, it means the current power has reached the maximum; when is 1 and Still less than , it means that the user currently does not accept the scheduling, that is, ;when is 0 and Still less than When , it means it can be dispatched; and in all time periods The sum of no more than , you can pass To adjust the frequency of power outages that users accept during the entire charging cycle, thereby controlling the overall satisfaction of users with their participation in scheduling behavior.

[0103] For reaching The rechargeable EV can be dispatched as both a dispatchable capacity resource and a dispatchable power resource. Its dispatchable capacity and power are:

[0104]

[0105] When controlling user discharge behavior, the satisfaction parameter is also introduced. and control factors :

[0106]

[0107] in, express Reached ; express Not reached But the user does not accept the scheduling, i.e. ; express Not reached But the user allows scheduling, i.e. ; and by limiting This method controls the maximum number of times a user can discharge and participate in scheduling in a day, thereby ensuring user experience and battery life.

[0108] In summary, by fine-tuning the dispatchable resource characteristics of the three types of EV users under different SOC conditions and introducing user satisfaction parameters to flexibly constrain the dispatch times, we can fully tap the regulation potential of electric vehicle clusters at the substation level while ensuring the basic energy needs of users, and realize controllable and flexible resource aggregation and dynamic response scheduling.

[0109] In this embodiment, by dynamically determining the individual feasible domain based on the current power level of the electric vehicle and the dispatch type, and limiting the dispatch capacity in combination with satisfaction parameters, it is possible to ensure that each electric vehicle can fully participate in the substation dispatch process without affecting the basic travel needs of users, thereby improving the robustness and flexibility of the dispatch strategy. At the same time, by limiting the feasible domain to the actual available power and capacity range, it can effectively prevent the dispatch plan from exceeding the grid's carrying capacity, reduce the impact of load fluctuations, thereby improving stability and user satisfaction, and strengthening the ability of electric vehicle clusters to participate in grid dispatch as elastic loads and the control accuracy.

[0110] In one embodiment, the steps of obtaining the cluster feasible region of the electric vehicle cluster in the substation area by using Chino polyhedron modeling and Minkowski summation according to each individual feasible region include:

[0111] A Qino polyhedron model of each individual feasible region is constructed, and Minkowski summation is performed on each Qino polyhedron model to obtain the cluster feasible region.

[0112] In this embodiment, the computer first obtains the current state parameters of each electric vehicle from the EV cluster management system, including but not limited to state of charge, maximum / minimum charge and discharge power, and user satisfaction thresholds. Based on these parameters, the modeling module is invoked to construct an individual feasible domain for each EV, taking into account its type and current adjustable constraints. To achieve efficient and scalable modeling, this feasible domain is represented as a Chino polyhedron. The Chino polyhedron's polygonal boundaries in the power and capacity dimensions are defined by multiple generating vectors, forming a convex set geometry that accurately represents the flexible space for single-vehicle scheduling.

[0113] Next, the cluster feasible region construction process begins. This process involves the parallel computing module performing a Minkowski summation operation, collectively summing the Chino polyhedron models of multiple EVs, thereby overlaying the feasible solution regions of each individual in the mathematical space. This operation involves coordinate superposition calculations on the vector set, gradually generating the final cluster Chino polyhedron model. Given the potentially large number of clusters, this process utilizes a sparse vector optimization strategy and parallel batch summation to ensure scalability and real-time performance of the model merging process.

[0114] After completing the above set summation, the resulting cluster feasible region model can be used as input to the grid dispatch algorithm for subsequent implementation of strategies such as unified power allocation, peak-valley response, or ancillary service regulation. Notably, because each individual model includes adjustable ranges and non-adjustable limits, the resulting cluster feasible region is truly operational and accurately reflects the full adjustable space available for the EV cluster under the premise of safe operation during the current period, avoiding the scheduling risks caused by false capacity estimates.

[0115] In one example, the feasible domain of a single electric vehicle (EV) is first determined. Based on its charging time window, state parameters, and power constraints, a convex polyhedron representation is formed that encompasses all feasible charging and discharging behaviors. Based on the geometric characteristics of this feasible domain, a corresponding parameter vector and parameter matrix are constructed to describe its shape and boundaries.

[0116] Next, the normal vectors corresponding to each boundary plane of the convex polyhedron are extracted to generate a normal vector matrix. Furthermore, the extension distance of the polyhedron in the direction of each normal vector is calculated to obtain its geometric circumscribed properties. This step provides directional information and scale basis for the subsequent inscribed modeling of the zonotope.

[0117] On this basis, we search for a zonotope with the largest volume inside the convex polyhedron, so that it fits as tightly as possible within the feasible region of the EV. The zonotope is described by two or more generating vectors, and its center position and generating matrix can be used for further set operations.

[0118] For multiple individual EVs, a Minkowski sum can be performed on each inscribed zonotope. This involves superimposing the generating vectors of each zonotope with their center of symmetry in vector space. This operation directly superimposes the center of symmetry of each individual zonotope with the set of generating vectors, yielding the overall center of symmetry and range of the EV aggregate in space.

[0119] Finally, based on the aggregated calculation results, the feasible region of the EV cluster is determined. This cluster feasible region not only preserves the convexity and physical constraints of the scheduling boundary of a single EV, but also reflects the integration capability of multiple EVs under simultaneous scheduling, laying the data and geometric foundation for subsequent power optimization scheduling and robust constraint modeling at the substation level.

[0120] It can be understood that by constructing a Chino polyhedron model to represent the feasible domain of individual electric vehicles, all possible adjustment combinations of each individual under the current constraints can be accurately described in multidimensional space, which is more expressive than the traditional interval model; then, these individual models are unified and merged using the Minkowski summation method, which can fully preserve the adjustment boundary characteristics of each individual and achieve additive integration of dispatch capabilities. The above execution process can significantly improve the accuracy and flexibility of cluster modeling, making the power grid dispatch strategy more dynamic and adaptable; it can also support the real-time computing needs of large-scale EV cluster dispatch systems at multiple time scales and improve the response efficiency of the dispatch strategy. Therefore, this method is not only efficient in construction process and strong in expressiveness, but also has good system scalability and user-side acceptance, effectively enhancing the practicality and stability of EV clusters participating in power grid flexibility dispatch.

[0121] In order to achieve efficient modeling of the feasible domain of individual electric vehicles, the Zonotope model is first used to mathematically represent the feasible domain of a single EV. Specifically, the Zonotope model is based on the symmetry center. As the base point, multiple vectors are generated by linear combination With direction vector To construct a closed, symmetrical convex geometry, where the direction vector , represents the range of variation of all possible scheduling states in each dimension, and is the number of generators of the single EV model, that is, the zonotope mathematical model is .

[0122] In one embodiment, the expression of the Qino polyhedron model of each individual feasible region is:

[0123]

[0124] in, represents the center point vector of the i-th Qino polyhedron model, Indicates the end time. Indicates the initial time, Indicates the maximum power value. Indicates the minimum power value. represents the first generating vector of the i-th Qino polyhedron model, Indicates fast charging power. Represents the second generating vector of the i-th Qino polyhedron model.

[0125] The center points in the Zonotope model correspond to the midpoint of the EV scheduling time interval and the midpoint of the acceptable power interval, reflecting its average state within the scheduling window. Furthermore, to describe the geometric boundary of the feasible region, the model constructs two vectors: and , It mainly expresses the joint changes of scheduling time dimension and power boundary, It is used to reflect the SOC change caused by power regulation.

[0126] It can be seen that the feasible domain of a single EV is precisely constructed through the above two generating vectors, which fully considers multiple factors such as scheduling time, power limit and SOC constraint.

[0127] After completing the Zonotope modeling of all monomeric EVs, the feasible domain of the entire cluster is further constructed through the additive properties of the Zonotope structure. Specifically, the center point of the cluster The center point of all monomers The vector sum of , that is:

[0128]

[0129] The generated vector set of the cluster Then it is the concatenation result of all monomer generation vectors:

[0130]

[0131] The resulting cluster Zonotope model not only preserves the scheduling characteristics of each individual EV in terms of time, energy, and power, but also fully demonstrates the overall cluster's regulatory capacity through collective addition. This model possesses highly linear expression capabilities and can be directly embedded in subsequent modules such as optimization scheduling, serving as a crucial foundational data structure for electric vehicle clusters to interact with the grid.

[0132] In one embodiment, the expression of the robust optimization constraint is:

[0133]

[0134] in, represents the charging power of the i-th electric vehicle at time t, Indicates the safety margin of the station area, Indicates the total number of electric vehicles in the area. Indicates the rated power of the transformer in the substation area. represents the prediction error, Represents the predicted household load power.

[0135] Under this constraint, EV cluster scheduling can account for household load uncertainty while ensuring that all EV charging activities do not cause transformer overload, thereby improving the safety and stability of substation operations. Especially in the context of large-scale EV integration into the distribution network, this constraint provides a solid safety framework for scheduling algorithms by setting clear boundary conditions for EV scheduling capabilities.

[0136] Furthermore, before using this robust constraint, it is necessary to obtain the dispatchable capacity and dispatchable power of the substation at each moment based on the feasible domain data of the substation EV cluster. Specifically, the feasible domain is constructed based on the charging demand, SOC state, dispatching strategy and satisfaction parameters of each EV, and is aggregated into the joint feasible domain of the entire EV cluster through geometric methods. The dispatch boundary at each moment can be dynamically extracted from the feasible domain, that is, the dispatchable capacity and power are obtained, which are the above constraints. Provide contextual conditions to make it temporal and physically feasible.

[0137] Applying this constraint to the scheduling model can ensure that the total power distribution of the EV cluster is reasonably allocated within the transformer capacity, thereby avoiding power outages or equipment damage caused by overload; secondly, introducing a safety margin and prediction error factor allows the model to maintain operational stability in the face of unpredictable fluctuations in household electricity consumption, thereby improving the robustness and practicality of the scheduling scheme; finally, through dynamic boundary adjustment based on the feasible domain, it has stronger adaptability and intelligent response capabilities, providing theoretical support and technical paths for achieving more efficient distribution network-side energy management.

[0138] The following describes the multi-area electric vehicle cluster scheduling optimization device provided by the embodiment of the present application. The multi-area electric vehicle cluster scheduling optimization device described below and the multi-area electric vehicle cluster scheduling optimization method described above can be referenced to each other. Figure 2 As shown, the present application provides a multi-area electric vehicle cluster scheduling optimization device, the device comprising:

[0139] The individual type determination module 201 is used to determine the individual type of each electric vehicle in the multi-area electric vehicle cluster according to the vehicle parameters of each electric vehicle, and the individual types include non-dispatchable type, transferable type and rechargeable type;

[0140] The cluster feasible region determination module 202 is used to establish the individual feasible region of each electric vehicle based on the vehicle parameters and individual type of each electric vehicle, and obtain the cluster feasible region of the electric vehicle cluster in the substation by using Chino polyhedron modeling and Minkowski summation based on each individual feasible region;

[0141] The electric vehicle cluster scheduling module 203 is used to perform multi-station electric vehicle cluster scheduling in the cluster feasible domain based on preset robust optimization constraints. The robust optimization constraints are used to ensure that the station transformer is not overloaded when scheduling electric vehicle charging behavior, while retaining safety redundancy for extreme changes in household loads.

[0142] In one embodiment, the vehicle parameters include the current power level, the expected power level, the rated charging power, the charging power, and the discharging power; the individual type determination module 201 includes:

[0143] a first individual type determination unit, configured to, for each electric vehicle, determine that if the current power value of the electric vehicle is not less than the corresponding expected power value and the charging power of the electric vehicle is equal to the corresponding rated charging power, then the individual type of the electric vehicle is a non-dispatchable type;

[0144] an electric vehicle discharge determination unit, configured to determine whether the electric vehicle is prohibited from discharging if the charging power of the electric vehicle is an adjustable charging power;

[0145] The second individual type determining unit is configured to determine that the individual type of the electric vehicle is a transferable type when the electric vehicle is prohibited from discharging, and to determine that the individual type of the electric vehicle is a chargeable and dischargeable type when the electric vehicle is allowed to discharge.

[0146] In one embodiment, the unschedulable expression is:

[0147]

[0148] The transferable expression is:

[0149]

[0150] The expression for the rechargeable and dischargeable type is:

[0151]

[0152] in, represents the charging power of the non-dispatchable electric vehicle at time t, Indicates the rated charging power of the electric vehicle, represents the power value of the electric vehicle at time t in the non-dispatchable type, Indicates the expected power value, Indicates the maximum power value. represents the power value of the electric vehicle at time t0 in the non-dispatchable type, Indicates the charging efficiency, Indicates the charging time interval, represents the charging power of the transferable electric vehicle at time t, Indicates fast charging power. Indicates slow charging power. represents the power value of the transferable electric vehicle at time t, Indicates the minimum power value. represents the power value of the transferable electric vehicle at time t0, represents the charging power of the rechargeable and dischargeable electric vehicle at time t, Indicates the discharge power, Indicates the power value of a rechargeable and dischargeable electric vehicle at time t, Indicates the power value of a rechargeable and dischargeable electric vehicle at time t0, represents the discharge efficiency, represents the dispatchable power, Indicates the discharge capacity of a rechargeable electric vehicle.

[0153] In one embodiment, the vehicle parameters include a current power value and a minimum power value; the cluster feasible region determination module 202 includes:

[0154] The first unit for determining a feasible region of an electric vehicle is configured to determine a feasible region of an electric vehicle of a non-dispatchable type, wherein the feasible region of the electric vehicle is a non-dispatchable capacity and a non-dispatchable power.

[0155] A second individual feasible domain determining unit is configured to determine, for an electric vehicle of transferable individual type, that when the current power value of the electric vehicle does not reach the corresponding minimum power value, the individual feasible domain of the electric vehicle is non-dispatchable capacity and non-dispatchable power; and when the current power value of the electric vehicle reaches the corresponding minimum power value, the individual feasible domain of the electric vehicle is only dispatchable power and the dispatchable capacity is zero, wherein the dispatchable power in the individual feasible domain of the electric vehicle is controlled by a preset first satisfaction parameter;

[0156] The third individual feasible domain determination unit is used for an electric vehicle whose individual type is a rechargeable and dischargeable type. When the current power value of the electric vehicle does not reach the corresponding minimum power value, the individual feasible domain of the electric vehicle is an undispatched capacity and an undispatched power. When the current power value of the electric vehicle reaches the corresponding minimum power value, the individual feasible domain of the electric vehicle is a dispatchable power and a dispatchable capacity, wherein the dispatchable power and the dispatchable capacity are controlled by a preset second satisfaction parameter in the individual feasible domain of the electric vehicle.

[0157] In one embodiment, the cluster feasible region determination module 202 includes:

[0158] The cluster feasible region determination unit is used to construct a Chino polyhedron model of each individual feasible region and perform Minkowski summation on each Chino polyhedron model to obtain the cluster feasible region.

[0159] In one embodiment, the expression of the Qino polyhedron model of each individual feasible region is:

[0160]

[0161] in, represents the center point vector of the i-th Qino polyhedron model, Indicates the end time. Indicates the initial time, Indicates the maximum power value. Indicates the minimum power value. represents the first generating vector of the i-th Qino polyhedron model, Indicates fast charging power. Represents the second generating vector of the i-th Qino polyhedron model.

[0162] In one embodiment, the expression of the robust optimization constraint is:

[0163]

[0164] in, represents the charging power of the i-th electric vehicle at time t, Indicates the safety margin of the station area, Indicates the total number of electric vehicles in the area. Indicates the rated power of the transformer in the substation area. represents the prediction error, Represents the predicted household load power.

[0165] In one embodiment, the present application also provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the multi-station electric vehicle cluster scheduling optimization method as described in any one of the above embodiments.

[0166] In one embodiment, the present application also provides a computer device having computer-readable instructions stored therein. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the multi-station electric vehicle cluster scheduling optimization method as described in any one of the above embodiments.

[0167] Schematically, as Figure 3 As shown, Figure 3 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. The computer device 300 can be provided as a server. Figure 3 Computer device 300 includes a processing component 302, which further includes one or more processors, and a memory resource represented by memory 301 for storing instructions executable by processing component 302, such as an application. The application stored in memory 301 may include one or more modules, each corresponding to a set of instructions. In addition, processing component 302 is configured to execute the instructions to perform the multi-station electric vehicle cluster scheduling optimization method according to any of the above embodiments.

[0168] The computer device 300 may further include a power supply component 303 configured to perform power management of the computer device 300, a wired or wireless network interface 304 configured to connect the computer device 300 to a network, and an input / output (I / O) interface 305. The computer device 300 may operate based on an operating system stored in the memory 301, such as Windows Server™, Mac OS X™, Unix™, Linux™, Free BSD™, or the like.

[0169] Those skilled in the art will understand that Figure 3The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0170] Finally, it should be noted that, in this article, relational terms such as first and second are merely used to distinguish one entity or operation from another, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprise," "include," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. Without further restriction, an element defined by the phrase "comprising a..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element. Herein, "one," "said," "the," and "its" may also include plural forms unless the context clearly indicates otherwise. A plurality refers to at least two, such as 2, 3, 5, or 8. "And / or" includes any and all combinations of the relevant listed items.

[0171] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referenced to each other.

[0172] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multi-area electric vehicle cluster scheduling optimization method, characterized in that: The method comprises: Determining, based on vehicle parameters of each electric vehicle, an individual type of each electric vehicle in a multi-area electric vehicle cluster, wherein the individual type includes a non-dispatchable type, a transferable type, and a rechargeable and dischargeable type; Based on the vehicle parameters and individual type of each electric vehicle, an individual feasible region of each electric vehicle is established, and according to each individual feasible region, a cluster feasible region of the electric vehicle cluster in the station area is obtained by using Chino polyhedron modeling and Minkowski summation; Based on preset robust optimization constraints, multi-station electric vehicle cluster scheduling is performed in the cluster feasible domain. The robust optimization constraints are used to ensure that the station transformer is not overloaded when scheduling electric vehicle charging behavior, while retaining safety redundancy for extreme changes in household loads.

2. The multi-area electric vehicle cluster scheduling optimization method according to claim 1 is characterized in that: The vehicle parameters include a current power value, an expected power value, a rated charging power, a charging power, and a discharging power; and the step of determining the individual type of each electric vehicle in a multi-area electric vehicle cluster based on the vehicle parameters of each electric vehicle includes: For each of the electric vehicles, if the current power value of the electric vehicle is not less than the corresponding expected power value, and the charging power of the electric vehicle is equal to the corresponding rated charging power, then the individual type of the electric vehicle is non-dispatchable; If the charging power of the electric vehicle is an adjustable charging power, determining whether the electric vehicle is prohibited from discharging; When the electric vehicle is prohibited from discharging, the individual type of the electric vehicle is a transferable type, and when the electric vehicle is allowed to discharge, the individual type of the electric vehicle is a chargeable and dischargeable type.

3. The multi-area electric vehicle cluster scheduling optimization method according to claim 2 is characterized in that: The expression of the unschedulable type is: The transferable expression is: The expression of the rechargeable and dischargeable type is: in, represents the charging power of the electric vehicle at time t in the non-dispatchable type, Indicates the rated charging power of the electric vehicle, represents the power value of the electric vehicle at time t in the non-dispatchable type, Indicates the expected power value, Indicates the maximum power value. represents the power value of the electric vehicle at time t0 in the non-dispatchable type, Indicates the charging efficiency, Indicates the charging time interval, represents the charging power of the transferable electric vehicle at time t, Indicates fast charging power. Indicates slow charging power. represents the power value of the transferable electric vehicle at time t, Indicates the minimum power value. represents the power value of the transportable electric vehicle at time t0, represents the charging power of the rechargeable electric vehicle at time t, Indicates the discharge power, represents the power value of the rechargeable electric vehicle at time t, represents the power value of the rechargeable electric vehicle at time t0, represents the discharge efficiency, represents the dispatchable power, Indicates the discharge capacity of the rechargeable and dischargeable electric vehicle.

4. The multi-area electric vehicle cluster scheduling optimization method according to claim 1, characterized in that: The vehicle parameters include a current power value and a minimum power value; the step of establishing an individual feasible domain for each electric vehicle based on the vehicle parameters and individual type of each electric vehicle includes: For electric vehicles with non-dispatchable individual types, the individual feasible region of the electric vehicle is non-dispatchable capacity and non-dispatchable power; For an electric vehicle of transferable individual type, when the current power value of the electric vehicle does not reach the corresponding minimum power value, the individual feasible domain of the electric vehicle is non-dispatchable capacity and non-dispatchable power. When the current power value of the electric vehicle reaches the corresponding minimum power value, the individual feasible domain of the electric vehicle is only dispatchable power and the dispatchable capacity is zero. The dispatchable power in the individual feasible domain of the electric vehicle is controlled by a preset first satisfaction parameter. For an electric vehicle whose individual type is rechargeable and dischargeable, when the current power value of the electric vehicle does not reach the corresponding minimum power value, the individual feasible domain of the electric vehicle is non-dispatchable capacity and non-dispatchable power; when the current power value of the electric vehicle reaches the corresponding minimum power value, the individual feasible domain of the electric vehicle is dispatchable power and dispatchable capacity, wherein the dispatchable power and dispatchable capacity in the individual feasible domain of the electric vehicle are controlled by a preset second satisfaction parameter.

5. The multi-area electric vehicle cluster scheduling optimization method according to claim 1, characterized in that: The step of obtaining the cluster feasible region of the electric vehicle cluster in the substation area by using Chino polyhedron modeling and Minkowski summation according to each of the individual feasible regions includes: A Chino polyhedron model of each individual feasible region is constructed, and Minkowski summation is performed on each Chino polyhedron model to obtain the cluster feasible region.

6. The multi-area electric vehicle cluster scheduling optimization method according to claim 5, characterized in that: The expression of the Qino polyhedron model of each individual feasible region is: in, represents the center point vector of the i-th Qino polyhedron model, Indicates the end time. Indicates the initial time, Indicates the maximum power value. Indicates the minimum power value. represents the first generating vector of the i-th Qino polyhedron model, Indicates fast charging power. Represents the second generating vector of the i-th Qino polyhedron model.

7. The multi-area electric vehicle cluster scheduling optimization method according to claim 1, characterized in that: The expression of the robust optimization constraint condition is: in, represents the charging power of the i-th electric vehicle at time t, Indicates the safety margin of the station area, Indicates the total number of electric vehicles in the area. Indicates the rated power of the transformer in the substation area. represents the prediction error, Represents the predicted household load power.

8. A multi-area electric vehicle cluster dispatch optimization device, characterized in that: The device comprises: An individual type determination module is used to determine the individual type of each electric vehicle in a multi-area electric vehicle cluster according to the vehicle parameters of each electric vehicle, wherein the individual type includes a non-dispatchable type, a transferable type, and a rechargeable and dischargeable type; A cluster feasible region determination module is used to establish an individual feasible region for each electric vehicle based on the vehicle parameters and individual type of each electric vehicle, and to obtain a cluster feasible region for the electric vehicle cluster in the substation by using Chino polyhedron modeling and Minkowski summation according to each individual feasible region; The electric vehicle cluster scheduling module is used to perform multi-station electric vehicle cluster scheduling in the cluster feasible domain based on preset robust optimization constraints. The robust optimization constraints are used to ensure that the station transformer is not overloaded when scheduling electric vehicle charging behavior, while retaining safety redundancy for extreme changes in household load.

9. A storage medium, characterized in that: The storage medium stores computer-readable instructions, which, when executed by one or more processors, enable the one or more processors to execute the steps of the multi-area electric vehicle cluster scheduling optimization method as described in any one of claims 1 to 7.

10. A computer device, characterized in that: include: one or more processors, and memory; The memory stores computer-readable instructions, and when the computer-readable instructions are executed by the one or more processors, the steps of the multi-area electric vehicle cluster scheduling optimization method as described in any one of claims 1 to 7 are executed.

Images