A charging and discharging control method, system, device and medium of an electric vehicle

Abstract

The application discloses a kind of electric vehicle's charge-discharge control method, system, equipment and medium, it is characterized in that, including: obtaining the charge-discharge time period data and battery state data of electric vehicle, and the topology structure parameter and power load data of three-phase ac-dc distribution network;Each electric vehicle is clustered, and the electric vehicle cluster group is obtained;Topology structure parameter, power load data and initial charge-discharge power are input to power flow equation, and the power grid power flow data is obtained, the initial battery state of charge value and target battery state of charge value of each electric vehicle cluster group are input to optimization model, to minimize network loss cost function, electric vehicle charge-discharge cost and voltage deviation function are solved to optimization model as optimization target, and target charge-discharge power is obtained;Each electric vehicle is controlled based on each target charge-discharge power and charges-discharges.The present application can inhibit three-phase ac-dc distribution network node voltage out-of-limit, reduce network loss cost and charging cost.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle charging technology, and in particular to a charging and discharging control method, system, device and medium for electric vehicles. Background Technology

[0002] With the rapid growth of the electric vehicle market share, large-scale electric vehicle integration into the power distribution network has become the norm, and their charging and discharging behavior has a significant impact on the power grid load distribution and operating status. At the same time, the power distribution network is gradually developing towards a three-phase AC / DC hybrid structure, which can flexibly integrate electric vehicles and distributed energy resources, but also makes the power grid operation mechanism more complex.

[0003] However, existing technologies often employ a method of modeling and centrally optimizing control for each individual electric vehicle. When large-scale electric vehicle integration is implemented, this can lead to excessively high dimensionality and computational complexity in the control model, making it difficult to obtain effective control results within a limited timeframe. This, in turn, affects the real-time performance and efficiency of charge and discharge control. Furthermore, in a three-phase AC / DC distribution network environment, existing methods do not adequately consider the coordinated effects of grid power flow distribution, voltage constraints, and converter operating characteristics. This makes it difficult to balance the charging and discharging needs of electric vehicles with grid operational constraints, resulting in low overall charge and discharge control efficiency and failing to meet practical application requirements. Summary of the Invention

[0004] This invention provides a charging and discharging control method, system, device, and medium for electric vehicles, which can improve the efficiency of charging and discharging control for large-scale electric vehicles in a three-phase AC / DC power distribution network.

[0005] In a first aspect, embodiments of the present invention provide a charging and discharging control method for an electric vehicle, comprising: Acquire data on the charging and discharging time periods and battery status of several electric vehicles, as well as the topology parameters and power load data of the three-phase AC / DC distribution network; Based on the charging and discharging time period data and the battery state data, the electric vehicles are clustered to obtain several electric vehicle clusters; The topology parameters, the power load data, and the initial charging and discharging power of each electric vehicle are input into a preset power flow equation to obtain the power flow data of the three-phase AC / DC distribution network. The power flow data, the initial battery state of charge value and the target battery state of charge value of each electric vehicle cluster are input into a preset optimization model. The optimization model is solved with the goal of minimizing the network loss cost function, the electric vehicle charging and discharging cost function and the voltage deviation function to obtain the target charging and discharging power of each electric vehicle cluster. The electric vehicles are charged and discharged based on the target charging and discharging power.

[0006] This invention provides a comprehensive data foundation for subsequent optimization calculations by achieving a complete characterization of electric vehicle charging and discharging behavior and grid operating status, thereby reducing the impact of information uncertainty on control decisions and improving the accuracy and efficiency of large-scale electric vehicle charging and discharging control in three-phase AC / DC distribution networks. By equivalently aggregating large-scale electric vehicles, the dimensionality of the optimization problem is significantly reduced, computational complexity is decreased, and thus the optimization solution speed is improved, increasing the computational efficiency of large-scale electric vehicle charging and discharging control in three-phase AC / DC distribution networks. By implementing multi-variable collaborative optimization based on grid operating constraints and electric vehicle charging demands, distribution network losses are reduced and voltage deviations are suppressed, improving system operating economy and stability, thereby enhancing the overall optimization efficiency of large-scale electric vehicle charging and discharging control in three-phase AC / DC distribution networks. Finally, by converting optimization results into actual control commands, coordinated regulation of electric vehicles and AC / DC distribution networks is achieved, thereby improving grid operating performance while meeting user charging needs and increasing the execution efficiency of large-scale electric vehicle charging and discharging control in three-phase AC / DC distribution networks.

[0007] Furthermore, based on the charging / discharging time period data and the battery state data, the electric vehicles are clustered to obtain several electric vehicle clusters, including: Extract the parking charging time of each electric vehicle from the charging and discharging time period data, and extract the initial state of charge value and target state of charge value of each electric vehicle from the battery state data. Based on the parking charging time, the initial state of charge value, and the target state of charge value, the electric vehicles are clustered to merge electric vehicles of the same type that have similar parking charging times, the same initial state of charge value and target state of charge value, and similar acceptable depth of discharge, thus obtaining several electric vehicle clusters.

[0008] This invention extracts the parking charging time, initial state of charge value, and target state of charge value of electric vehicles, and merges vehicles with similar characteristics into clusters, which greatly reduces the dimensionality of decision variables in scheduling optimization, improves the convergence speed and solution efficiency of optimization calculation, and thus improves the charging and discharging control efficiency of large-scale electric vehicles in three-phase AC / DC distribution networks.

[0009] Furthermore, the step of inputting the topology parameters, the electrical load data, and the initial charging and discharging power of each electric vehicle into a preset power flow equation to obtain the power flow data of the three-phase AC / DC distribution network includes: The initial injection current and initial node voltage of each node in the three-phase AC / DC distribution network are determined based on the topology parameters, the power load data, and the initial charging and discharging power. Substitute each of the initial injection currents and each of the initial node voltages into the power flow equations, and iteratively solve the power flow equations using a preset Newton iteration method. In each iteration, calculate the current imbalance of each node, and correct each of the initial node voltages based on the current imbalance to obtain the target node voltage. The injected current of each node in the three-phase AC / DC distribution network is calculated based on the target node voltage, and it is determined whether the node current imbalance of each injected current meets the preset convergence condition. If it meets the condition, the iteration stops, and the power flow data of the three-phase AC / DC distribution network is output. The power flow data includes the voltage, current and power distribution results of each node.

[0010] This invention uses the Newton-Raphson iterative method to accurately solve the AC / DC coupled power flow equations, obtaining accurate node voltage, current, and power distributions in a three-phase AC / DC distribution network. This provides a reliable grid state reference for subsequent electric vehicle charging and discharging optimization, thereby improving the charging and discharging control efficiency of large-scale electric vehicles in a three-phase AC / DC distribution network.

[0011] Furthermore, the process of constructing the power flow equations includes: Based on the topology parameters, construct the node admittance matrix of the three-phase AC power grid and the node admittance matrix of the DC power grid, and determine the power flow equations of the three-phase AC power grid and the DC power grid respectively based on the node admittance matrix of the three-phase AC power grid and the node admittance matrix of the DC power grid. By combining the preset three-phase voltage source converter model, the preset DC converter model, the three-phase AC power flow equation, and the DC power flow equation, a power flow equation including the three-phase AC power grid, the DC power grid, and the AC / DC conversion device is obtained.

[0012] This invention establishes a complete power flow equation for a three-phase AC / DC hybrid distribution network by constructing a unified admittance matrix for both the three-phase AC and DC power grids and introducing steady-state models for voltage source converters and DC converters. This achieves an accurate characterization of the AC / DC coupling characteristics, thereby improving the charging and discharging control efficiency for large-scale electric vehicles in a three-phase AC / DC distribution network.

[0013] Furthermore, the optimization model is solved by minimizing the network loss cost function, the electric vehicle charging and discharging cost function, and the voltage deviation function as the optimization objective functions to obtain the target charging and discharging power of each electric vehicle cluster, including: Within the decision space of the optimization model, the charging and discharging power of each electric vehicle cluster, the voltage conversion ratio phasor of the three-phase voltage source converter model, and the duty cycle of the DC-DC converter model are determined as decision variables. The initial battery state of charge (SOC) value, target battery SOC value, and power flow data of the three-phase AC / DC distribution network of each electric vehicle cluster are input into the optimization model. The optimization objective function is to minimize the network loss cost function, the electric vehicle charging and discharging cost function, and the voltage deviation function. Under the preset set of constraints, the optimization model is solved by the preset optimization solution algorithm to obtain the optimal solution of each decision variable. Extract the target charge / discharge power of each electric vehicle cluster from the optimal solution.

[0014] This invention uses the charging and discharging power of electric vehicle clusters, VSC voltage conversion ratio, and DC / DC converter duty cycle as decision variables, with the optimization objective of minimizing network loss costs and voltage deviation. Under multiple constraints such as power flow, SOC, and voltage and current limits, the optimal charging and discharging power is solved, achieving coordinated optimization of economy and voltage quality. This improves the charging and discharging control efficiency of large-scale electric vehicles in three-phase AC / DC distribution networks.

[0015] Furthermore, the process of constructing the optimization objective function includes: The active power loss value and electricity price at each discrete time point within the charging and discharging time period of each electric vehicle are obtained, and the products of each active power loss value and each electricity price are accumulated and summed to obtain the network loss cost function. At each of the discrete time points, the node three-phase voltage, neutral point voltage and node voltage of each AC node in the three-phase AC / DC distribution network are obtained, and the deviations of each node three-phase voltage, each neutral point voltage and each node voltage from the preset rated voltage are calculated respectively. All the deviations are accumulated to obtain the voltage deviation function. The network loss cost function, the preset electric vehicle charging and discharging cost function, the voltage deviation function, and the preset weighting coefficients are weighted and summed to obtain the optimization objective function.

[0016] This invention uses a weighted sum of the network loss cost function, the electric vehicle charging and discharging cost function, and the voltage deviation function as the optimization objective. This balances the economic efficiency of power grid operation with voltage quality, enabling the optimization model to effectively suppress three-phase voltage and neutral point voltage deviations while reducing active power loss costs. This, in turn, improves the charging and discharging control efficiency of large-scale electric vehicles in three-phase AC / DC distribution networks.

[0017] Furthermore, the set of constraints includes: AC / DC power flow constraints, upper and lower limits of electric vehicle state of charge constraints, mutual exclusion constraints of electric vehicle charging and discharging, maximum depth of discharge constraints of electric vehicles, upper and lower limits of AC / DC node voltage constraints, upper limit constraints of AC line current, upper limit constraints of DC line current, and node injection current constraints.

[0018] The embodiments of the present invention construct multi-dimensional constraints including AC / DC power flow, electric vehicle SOC, charge / discharge mutual exclusion, discharge depth, node voltage, line current and injection current, to ensure that the optimized solution is feasible within the grid safety boundary and battery physical limitations, effectively avoiding the risks of voltage overrun, equipment overload and battery abuse, thereby improving the reliability and safety of charge and discharge control for large-scale electric vehicles in three-phase AC / DC distribution networks.

[0019] Secondly, embodiments of the present invention provide a charging and discharging control system for an electric vehicle, the system comprising: an acquisition module, a clustering module, a solution module, and a control module; The acquisition module is used to acquire charging and discharging time period data and battery status data of several electric vehicles, as well as topology parameters and power load data of the three-phase AC / DC distribution network. The clustering module is used to cluster each electric vehicle based on the charging and discharging time period data and the battery state data to obtain several electric vehicle clusters. The solution module is used to input the topology parameters, the power load data, and the initial charging and discharging power of each electric vehicle into a preset power flow equation to obtain the power flow data of the three-phase AC / DC distribution network. The power flow data, the initial battery state of charge value and the target battery state of charge value of each electric vehicle cluster are input into a preset optimization model. The optimization model is solved with the goal of minimizing the network loss cost function, the electric vehicle charging and discharging cost function and the voltage deviation function to obtain the target charging and discharging power of each electric vehicle cluster. The control module is used to control the charging and discharging of each of the electric vehicles based on the target charging and discharging power.

[0020] This invention, through the collaborative work of the acquisition module, clustering module, solution module, and control module, first clusters to reduce dimensionality and then solves the AC / DC power flow and optimization model, realizing the equivalent aggregation and rolling optimization control of large-scale electric vehicles. This effectively reduces computational complexity, ensures the safe and economical operation of the power grid, and thus improves the efficiency and reliability of charging and discharging control of large-scale electric vehicles in three-phase AC / DC distribution networks.

[0021] Thirdly, embodiments of the present invention provide a terminal device, including: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus; The memory is used to store at least one executable instruction that causes the processor to perform the operation of a charging and discharging control method for an electric vehicle as described in this application.

[0022] Fourthly, embodiments of the present invention provide a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device or system where the computer-readable storage medium is located to perform a charging and discharging control method for an electric vehicle as described in this application.

[0023] The above description is merely an overview of the technical solutions of the embodiments of the present invention. In order to better understand the technical means of the embodiments of the present invention and to implement them in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the embodiments of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0024] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0025] Figure 1 This is a schematic flowchart of an embodiment of the charging and discharging control method for an electric vehicle provided in this application; Figure 2 This is a flowchart illustrating steps S201 to S202 provided in this application; Figure 3 This is a flowchart illustrating steps S301 to S303 provided in this application; Figure 4 This is a flowchart illustrating steps S401 to S403 provided in this application; Figure 5 This is a schematic diagram of an embodiment of the charging and discharging control method for an electric vehicle provided in this application. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0027] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0028] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0029] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0030] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0031] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0032] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0033] With the rapid growth of the electric vehicle market share, large-scale electric vehicle integration into the power distribution network has become commonplace, and their charging and discharging behavior has a significant impact on the grid load and operating status. Simultaneously, the power distribution network is gradually evolving into a three-phase AC / DC hybrid structure, which, while allowing for flexible integration of electric vehicles and distributed energy resources, also makes the operating mechanism more complex. However, existing technologies often employ methods of modeling and centrally optimizing control for each individual electric vehicle. In the case of large-scale integration, this can easily lead to excessively high model dimensionality and computational load, making it difficult to obtain effective control results within a limited time, thus affecting the real-time performance and efficiency of control. Furthermore, existing methods do not adequately consider the power flow distribution, voltage constraints, and converter operating characteristics of the three-phase AC / DC power distribution network, making it difficult to balance the charging and discharging needs of electric vehicles with grid operating constraints. This results in low overall charging and discharging control efficiency, failing to meet practical application requirements.

[0034] See Figure 1 In order to improve the efficiency of charging and discharging control of large-scale electric vehicles in a three-phase AC / DC power distribution network, an embodiment of the present invention provides a charging and discharging control method for electric vehicles, including steps S101 to S104. Step S101: Obtain charging and discharging time period data and battery status data of several electric vehicles, as well as topology parameters and power load data of the three-phase AC / DC distribution network. In some embodiments, firstly, the basic information of all electric vehicles applying to participate in the next-day charging and discharging service is obtained through the electric vehicle charging management platform. This basic information includes fixed parameters such as vehicle model, battery capacity, rated charging and discharging power, charging efficiency, discharging efficiency, and self-discharge rate. Simultaneously, the next-day charging and discharging time period data for each electric vehicle is obtained. This data includes the next-day charging and discharging start time and end time input by the user through a mobile terminal application; if the user does not actively input this data, it is predicted based on the average charging and discharging start and end times of the vehicle over the past 90 days. Furthermore, the current battery state of charge (SOC), target SOC, maximum acceptable depth of discharge (DOD), and historical default rate information for each electric vehicle are also obtained. Based on battery status (e.g., SOC not lower than the minimum allowable value, battery health meeting requirements) and default rate (e.g., historical default count less than a set threshold), a set of vehicles actually eligible to participate in the charging and discharging service is selected, and vehicles that do not meet the conditions are removed. Secondly, the topology parameters of the three-phase AC / DC distribution network are acquired in real time through the Supervisory Control and Data Acquisition (SCADA) system. These parameters include: the three-phase and neutral point admittance matrices of each node in the AC network, and the phase-to-phase and relative neutral point resistance, reactance, and susceptance parameters of each AC line; the admittance matrix of each node in the DC network, and the resistance parameters of each DC line; the connection node, capacity, control mode (master-slave control, voltage margin control, or droop control) and equivalent admittance of the three-phase voltage source converter (VSC); the connection node, type, equivalent conductance, efficiency curve, and control mode (constant voltage control or constant duty cycle control) of the DC-DC converter (Buck converter, Boost converter); and the location, tap range, and voltage regulation method of the transformer and traditional voltage regulating equipment. Next, the electricity load data for each time slot of the three-phase AC / DC distribution network on the following day is obtained. This data includes predicted active and reactive power values ​​for constant impedance (Z), constant current (I), constant power (P), and three types of loads (ZIP loads) at each AC node (excluding electric vehicles), as well as predicted constant power load values ​​for each DC node (excluding electric vehicles). Load forecasting is generated using a time-series forecasting method based on historical load data for the same period, weather factors, and date type (weekday or holiday), with a time resolution of 15 minutes. Simultaneously, time-of-use electricity price data for each time slot on the following day is obtained, including prices for peak, off-peak, and low-peak periods. Finally, the period from 00:00 to 24:00 on the following day is discretized into several equal-length time slots, with a time slot length of... A 15-minute timeframe is used, resulting in 96 time slots. Based on the next day's charging and discharging times for all selected vehicles, a set of vehicles in a parked charging state is generated within each time slot. For each electric vehicle, the number of time slots required for charging is calculated based on its required charge (calculated by multiplying the difference between the target SOC and the initial SOC by the battery capacity) and maximum charging power, and these time slots are allocated to the corresponding time slot sets. The calculation of required charge also considers a comprehensive correction based on the vehicle's battery-driven ratio, driving distance, and all-electric feasible distance for the day to improve the accuracy of charging demand prediction.

[0035] Through the above steps, basic data such as the charging and discharging time period of electric vehicles, battery status, distribution network topology and power load are obtained comprehensively and accurately. Based on historical behavior and real-time information, filtering and prediction are performed, providing a reliable data foundation for subsequent electric vehicle clustering, power flow calculation and rolling optimization. This enables the accuracy and efficiency of large-scale electric vehicle charging and discharging control in three-phase AC and DC distribution networks.

[0036] Step S102: Based on the charging and discharging time period data and the battery state data, the electric vehicles are clustered to obtain several electric vehicle clusters; Please refer to Figure 2 In some embodiments, the step of clustering each electric vehicle based on each charging and discharging time period data and each battery state data to obtain several electric vehicle clusters includes: steps S201 to S202. Step S201: Extract the parking charging time of each electric vehicle from the charging and discharging time period data, and extract the initial state of charge value and target state of charge value of each electric vehicle from the battery state data. In some embodiments, for each selected electric vehicle, the user-inputted start and end times for the next day's charging / discharging are extracted from the charging / discharging time period data; if the user does not input any, the average of the vehicle's historical 90-day charging / discharging start and end times is extracted as a substitute. Based on the charging / discharging start and end times, the total parking and charging time of the vehicle is calculated. Based on the battery state data, the current state of charge (SBC) of the vehicle is extracted as the initial SBC value. Extract the user-defined desired state of charge as the target state of charge value. If the user does not specify the expected state of charge, the default setting is 0.9 (i.e., 90%). The maximum acceptable depth of discharge for the vehicle is also extracted. Total battery capacity (Unit: kWh), Rated charge / discharge power (Unit: kW), Charging efficiency Discharge efficiency and self-discharge rate Based on the initial state of charge (SOC), target SOC, battery capacity, and charging efficiency, calculate the net amount of charge required to charge the vehicle from its initial SOC to the expected SOC. After considering charging efficiency, the actual amount of electricity that needs to be input from the grid is... Then, based on the vehicle's rated charging power... Calculate the theoretical minimum charging time while parked: The theoretical minimum parking charging time is compared with the total parking charging time. Comparison: If If the vehicle fails to meet the expected charging target within the current parking cycle, it will be removed from the current optimization cycle, or the user will be prompted to adjust the expected SOC. Then the parking and charging time of the vehicle will be recorded as... This information is used for subsequent cluster analysis. In addition, the vehicle type (model) information is extracted, including models I, II, III, IV, etc., with different models corresponding to different battery capacities and rated power parameters. Vehicles of the same model are grouped into the same model set. This is to facilitate subsequent clustering within the same vehicle model.

[0037] Step S202: Based on each of the parking charging times, each of the initial state of charge values, and each of the target state of charge values, the electric vehicles are clustered to merge electric vehicles of the same type that have similar parking charging times, the same initial state of charge values ​​and target state of charge values, and similar acceptable depths of discharge, to obtain several electric vehicle clusters.

[0038] In some embodiments, the set of electric vehicles participating in charging and discharging is assumed to be Each electric vehicle It has the following attributes: vehicle model ( (for all vehicle models), grid connection time Time away from the power grid Initial state of charge Target state of charge Battery capacity Maximum acceptable depth of discharge Rated charging power Charging efficiency Discharge efficiency First, all electric vehicles are categorized according to vehicle model. For each vehicle model... All electric vehicles of that model that require charging are considered as the set to be clustered for that model. Then, for each model Electric vehicles within the grid are merged according to the following rules: Time proximity judgment: The grid connection time and grid disconnection time of electric vehicles are discretized into time periods of a certain duration (e.g., one hour). For any two electric vehicles... If their grid connection times differ by no more than 1 unit of time, and their grid disconnection times differ by no more than 1 unit of time, then they are considered to have "similar parking and charging times." Initial SOC and target SOC similarity judgment: If the initial SOC values ​​of the two electric vehicles are equal (i.e., ... And the target SOC values ​​are equal (i.e.) If they are already classified by vehicle type, then the sameness condition is met. Same type judgment: Since they have already been classified by vehicle type... Grouping electric vehicles within the same group naturally satisfies the "same type" condition. Each cluster... As a whole, it participates in subsequent rolling optimization scheduling, and its charging and discharging power and To optimize the model's control variables, the above clustering equivalence processing simplifies the original scheduling problem of tens of thousands of electric vehicles into a scheduling problem of hundreds of clusters, effectively avoiding the curse of dimensionality and meeting the time requirements for online optimization computation.

[0039] Through the above steps, a large number of electric vehicles are clustered according to three feature dimensions: parking and charging time, initial SOC, and expected SOC, and merged into equivalent clusters. This significantly reduces the dimensionality of decision variables for optimization scheduling, effectively avoids the "curse of dimensionality" problem, and improves computational efficiency and convergence speed. As a result, it is possible to meet the time requirements for large-scale online rolling optimization of electric vehicles in a three-phase AC / DC distribution network.

[0040] Step S103: Input the topology parameters, the power load data, and the initial charging and discharging power of each electric vehicle into a preset power flow equation to obtain the power flow data of the three-phase AC / DC distribution network. Input the power flow data, the initial battery state of charge value and the target battery state of charge value of each electric vehicle cluster into a preset optimization model. Solve the optimization model with the goal of minimizing the network loss cost function, the electric vehicle charging and discharging cost function and the voltage deviation function to obtain the target charging and discharging power of each electric vehicle cluster. Please refer to Figure 3 In some embodiments, the step of inputting the topology parameters, the power load data and the initial charging and discharging power of each electric vehicle into a preset power flow equation to obtain the power flow data of the three-phase AC / DC distribution network includes: steps S301 to S303. Step S301: Determine the initial injection current and initial node voltage of each node in the three-phase AC / DC distribution network based on the topology parameters, the power load data, and the initial charging and discharging power. In some embodiments, firstly, the three-phase and neutral point admittance matrices of each node of the AC power grid are extracted from the topology parameters. For communication nodes and The admittance matrix between them has the following specific form: ; in, For communication nodes With nodes The 4×4 admittance matrix between the three phases (a, b, c) and the neutral point (N) includes mutual admittance and self-admittance. The first in the matrix line, number The elements of a column represent nodes. of AND node of Mutual admittance between phases (when (Time is self-guided admittance), each Since it is a complex number, it can be decomposed into conductivity. and susceptance : Simultaneously, the nodal admittance matrix of each node in the DC power grid is extracted. And the AC-side equivalent admittance of each voltage source converter (VSC). Equivalent conductance of each Buck converter Equivalent conductance of each Boost converter Next, initialize the voltage of each node. For AC nodes, set the initial voltage values ​​for each phase (a, b, c) and the neutral point (N): the three-phase phase voltage amplitude is set to 1.0 pu (per unit), the phase angles are 0°, -120°, and +120° respectively, and the initial neutral point voltage is set to 0 pu. For DC nodes, set the initial voltage value to 1.0 pu. For VSC internal nodes and DC / DC converter internal nodes, the initial voltage value is set to 1.0 pu according to the converter turns ratio. Record these initial values ​​as... and Then, based on the power load data and the initial charging and discharging power, the initial injection current of each node is calculated. For AC nodes... Injection current of ZIP loads other than electric vehicles The current injected into the electric vehicle is calculated from the active and reactive power of the load and the initial values ​​of the node voltage. The initial charge and discharge power of each cluster of electric vehicles , The initial values ​​of the node voltages are calculated. The AC node injection current satisfies: For DC nodes Initial injection current This is obtained by dividing the DC load power and the electric vehicle charging / discharging power by the initial value of the node voltage. For connections to AC nodes... and DC node The three-phase current between VSCs and DC node current The following coupling relationship is satisfied: ; in, To inject DC from the VSC DC side to the DC node DC current (scalar, real number); Inject an AC node into the VSC AC side The current phasors (complex numbers) of phases a, b, c and neutral point N; The voltage conversion ratio phasor for each phase is initially set to 1.0∠0°; Equivalent admittance of the VSC AC side; for The conjugate of complex numbers; DC node Voltage amplitude (real number, per unit value); For communication nodes The voltage phasors (complex numbers) of phases a, b, c and neutral point N; superscript " " denotes a conjugate complex number. For a Buck converter (step-down converter), let it be connected to node..." and nodes Between, internal nodes are denoted as The steady-state model of the Buck converter is as follows: ; in, For time slots Inside, the Buck converter connects to the node. Active power on the side (real number, positive value indicates power flowing into the node) (Negative values ​​indicate outflow). Internal nodes of the Buck converter In the time slot Voltage (controlled by duty cycle): ); For the high-voltage side node of the Buck converter In the time slot Voltage (real number); Low-voltage side node of Buck converter In the time slot Voltage (real number); For time slots Internal nodes of the Buck converter Flow to low-pressure side nodes The current (real number) is calculated using the following formula: ; Power is generated by the node Flow to Node The converter efficiency (real number) in the buck direction ); Power is generated by the node Flow to Node The converter efficiency (real number) in the reverse boost direction ).

[0041] The steady-state model for a Boost converter is as follows: ; in, For time slots Inside, the Boost converter connects to the node. Active power (real number) on the side; For the internal nodes of the Boost converter In the time slot Voltage (controlled by duty cycle): ); For the low-voltage side node of the Boost converter In the time slot Voltage (real number); For the high-voltage side node of the Boost converter In the time slot Voltage (real number); For time slots Internal nodes of the Boost converter Flow to high-voltage side nodes The current (real number) is calculated using the following formula: ; Power is generated by the node Flow to Node The converter efficiency (real number) in the boost direction ); Power is generated by the node Flow to Node The converter efficiency (real number) in the reverse buck direction The initial injection current and initial node voltage of all the above nodes are saved as initial values ​​for power flow calculation.

[0042] Step S302: Substitute each of the initial injection currents and each of the initial node voltages into the power flow equations, and iteratively solve the power flow equations using a preset Newton iteration method. In each iteration, calculate the current imbalance of each node, and correct each of the initial node voltages based on the current imbalance to obtain the target node voltage. In some embodiments, the initial injection current and initial node voltage obtained in step S301 are substituted into the power flow equations. These power flow equations include: Three-phase AC power flow equations: for each AC node... time slot satisfy: ; in, For nodes The injection current complex vector (including phases a, b, c and neutral point). For nodes The voltage complex vector, For nodes and The admittance matrix between them (as shown in step S201); DC power flow equations: ,in, They are time slots The injected current vector, voltage vector, and node admittance matrix of the DC node; VSC coupling equations; Buck converter equations; Boost converter equations. These equations together constitute the constraints for node current balance. In the... The next iteration ( In this process, the first step is to estimate the voltage at the current node. The theoretical injection currents at AC and DC nodes are calculated using the three-phase AC power flow equations and DC power flow equations, respectively. Simultaneously, based on the current voltage, given load power, and electric vehicle charging / discharging power, the actual injected current is calculated. For VSC and DC / DC converters, the current injected into the node is calculated based on the current voltage and converter control parameters (voltage conversion ratio or duty cycle). Then, the first node current imbalance is calculated for each node: ,in This includes the current imbalance (real and imaginary parts) of each phase and neutral point at AC nodes, as well as the current imbalance (scalar) at DC nodes. Arranged in order into unbalanced vectors .like (i.e., the maximum absolute value of each imbalance quantity) is greater than the preset convergence accuracy (e.g., If pu), then proceed to the Newton correction step, forming the Jacobian matrix. This matrix consists of the partial derivatives of each power flow equation with respect to each node voltage (real and imaginary parts of AC node voltages, and DC node voltages). Solve the linear equation system: The voltage correction amount is obtained. ,in This includes corrections for the real and imaginary parts of the three-phase and neutral point voltages at each AC node, and corrections for the voltage amplitude at each DC node. The node voltages are updated using the following formula: For AC nodes, the updated voltage complex vector The updated voltage amplitude is used directly in the next iteration; for DC nodes, the updated voltage amplitude... Similarly, save these values. Also, if the voltage conversion ratio of the VSC or the duty cycle of the DC / DC converter are used as state variables in the power flow calculation, update them accordingly. Repeat the above iterative process until the maximum value of the current imbalance is less than the convergence accuracy; the node voltage obtained at this point is the target node voltage. .

[0043] Step S303: Calculate the injected current of each node in the three-phase AC / DC distribution network based on the target node voltage, and determine whether the node current imbalance of each injected current meets the preset convergence condition. If it meets the condition, stop the iteration and output the power flow data of the three-phase AC / DC distribution network. The power flow data includes the voltage, current and power distribution results of each node.

[0044] In some embodiments, after the Newton iteration in step S302 converges, the node voltage obtained from the last iteration is... This refers to the target node voltage that satisfies the power flow equations. To ensure the reliability of the convergence results, the injection current of each node is recalculated based on this target node voltage. Specifically, the target node voltage is substituted into the three-phase AC power flow equations, DC power flow equations, VSC coupling equations, Buck converter equations, and Boost converter equations to calculate the theoretical injection current of each node under the target voltage. Simultaneously, based on the target node voltage and the given load power and electric vehicle charging / discharging power (using the initial charging / discharging power of each cluster or the power given in the previous optimization round), the actual injected current is calculated. Then calculate the current imbalance at the second node: ,judge Is it less than or equal to the preset convergence precision (e.g.) If the condition is met, the power flow calculation is confirmed to have converged, and the iteration stops. If the condition is not met, return to step S302 and continue iterating with the current target node voltage as the initial value until the convergence condition is met. After convergence, output the power flow data of the three-phase AC / DC distribution network. The power flow data includes: the voltage complex vector (amplitude and phase angle) of each AC node (including three phases and the neutral point); the voltage amplitude of each DC node; the current (real part and imaginary part), active power and reactive power of each phase of each AC line; the current and active power of each DC line; and the active power of each phase on the AC side of each VSC. and reactive power The calculation formula is as follows: ; in, subscript , Superscript indicates the real and imaginary parts of voltage or current. Indicates the neutral point; node voltages, currents, and transmitted power within each Buck / Boost converter; and overall network active power loss. The aforementioned power flow data will serve as the initial state and constraints for subsequent electric vehicle cluster rolling charge-discharge optimization control, used to solve for the optimal charge-discharge power, VSC voltage conversion ratio, and DC / DC converter duty cycle cycle by cycle.

[0045] In some embodiments, the process of constructing the power flow equations includes: constructing a three-phase AC grid node admittance matrix and a DC grid node admittance matrix based on the topology parameters, and determining the three-phase AC grid power flow equations and the DC grid power flow equations based on the three-phase AC grid node admittance matrix and the DC grid node admittance matrix, respectively; and simultaneously solving a preset three-phase voltage source converter model, a preset DC converter model, the three-phase AC grid power flow equations, and the DC grid power flow equations to obtain power flow equations that include a three-phase AC grid, a DC grid, and an AC / DC conversion device.

[0046] In some embodiments, a three-phase AC grid node admittance matrix and a DC grid node admittance matrix are constructed based on the topology parameters, and the three-phase AC grid power flow equation and the DC grid power flow equation are determined based on the three-phase AC grid node admittance matrix and the DC grid node admittance matrix, respectively. Specifically, the following steps are taken: First, the line parameters and transformer parameters of each node of the AC grid are extracted from the topology parameters obtained from the SCADA system. For each AC line or transformer branch, the resistance, reactance, and susceptance parameters between each phase (a, b, c) and the neutral point (N) are obtained. Based on these parameters, the AC node... With nodes The 4×4 admittance matrix between Its specific form is as follows: ; Each element Since it is a complex number, it can be decomposed into conductivity. and susceptance : ;when hour, For nodes The self-admittance is equal to all connections to the node. The sum of branch admittances; when hour, For nodes and The mutual admittance between branches is calculated, and the negative branch admittance is taken. For a DC grid, the resistance parameters of each DC line are extracted from the topology parameters. For each DC line, its conductance is calculated. The nodal admittance matrix of a DC power grid It is A real matrix, where the off-diagonal elements diagonal elements Based on the aforementioned admittance matrix, the power flow equations for a three-phase AC power grid and a DC power grid are constructed respectively. Three-phase AC power flow equations: For time slots... Communication nodes Injection current complex vector (Including phases a, b, c, and the neutral point) should be equal to all adjacent nodes. The sum of currents coupled through the admittance matrix, i.e.: ,in, For communication nodes In the time slot The voltage complex vector (including phases a, b, c and neutral point). Given the total number of AC nodes, this equation is a complex equation, with each node providing four complex equations (corresponding to a, b, c, and N), equivalent to eight real equations. DC power flow equations: for time slots... DC node injected current vector With node voltage vector satisfy: ,in, For each DC node, a real equation is provided for the admittance matrix (real number).

[0047] In some embodiments, a preset three-phase voltage source converter model, a preset DC converter model, the three-phase AC power flow equations, and the DC power flow equations are simultaneously established to obtain power flow equations that include the three-phase AC power grid, the DC power grid, and the AC / DC conversion device. Specifically, for each connected AC node... and DC node The three-phase voltage source converter (VSC) uses the following coupling model. This model considers the three-phase currents on the AC side of the VSC. and DC side current With AC node voltage and DC node voltage Connecting them: ; in, For each phase, the ideal voltage conversion ratio phasor is... The AC-side admittance of the VSC is given by the superscript "*", which indicates the conjugate complex number. This equation unifies the AC and DC-side currents of the VSC and can be directly incorporated into the injected currents of the AC and DC nodes. Secondly, for each DC-DC converter, a corresponding steady-state model is adopted according to its type (Buck converter or Boost converter). Buck converter: Assume the converter is connected to the node... (High-voltage side) and nodes Between (low-pressure side) internal nodes are denoted as Its model equations are: ; in, Duty cycle (control variable) For converter conductance, and These represent the forward and reverse efficiencies, respectively. Boost converter: Assume the converter is connected to the node... (Low-pressure side) and nodes Between (high-voltage sides), the internal nodes are denoted as Its model equations are: ; in, Duty cycle (control variable) For converter conductance, and The forward and reverse efficiencies are respectively. Finally, all the above models are combined to form a complete power flow equation. This set of equations together constitutes the complete mathematical expression of the power flow equation describing the steady-state operating characteristics of a three-phase AC / DC distribution network. In this set of equations, the unknown variables include: the three-phase and neutral point voltages at each AC node (complex vectors), the voltages at each DC node (real numbers), and the voltage transformation ratio phasors of each VSC. The control variables are the duty cycle of each DC / DC converter and the charging / discharging power of each electric vehicle cluster. By solving this system of equations using the Newton-Raphson iterative method, the power flow distribution of the power grid can be obtained.

[0048] Please refer toFigure 4 In some embodiments, the step of solving the optimization model with minimizing the network loss cost function, the electric vehicle charging and discharging cost function and the voltage deviation function as the optimization objective function to obtain the target charging and discharging power of each electric vehicle cluster includes: steps S401 to S403. Step S401: In the decision space of the optimization model, the charging and discharging power of each electric vehicle cluster, the voltage conversion ratio phasor of the three-phase voltage source converter model, and the duty cycle of the DC converter model are determined as decision variables. In some embodiments, firstly, the decision space of the optimization model is determined. This decision space consists of all control variables that need to be optimized, including: the charging and discharging power of each electric vehicle cluster; for each cluster... ( , For model (a set of clusters) in each optimization time slot Within this framework, two decision variables are defined: charging power. : indicates at node Clusters accessed at the location In the time slot Charging power (real number, unit: kW, non-negative); Discharging power : indicates at node Clusters accessed at the location In the time slot The discharge power (real number, unit: kW, non-negative). Simultaneously, the charge / discharge mutual exclusion condition requires... This means that the same power group cannot be charged and discharged simultaneously in the same time slot. Voltage conversion ratio phasor of a three-phase voltage source converter (VSC): For each VSC (the first... Each VSC is connected to the communication node. and DC node (between), defining each phase voltage conversion ratio phasor This phasor is a complex number and can be represented as: ,in The amplitude (real number, per unit value). Let be the phase angle (real number, radians). In actual optimization, it is usually... The real and imaginary parts are used as decision variables, or their magnitude and phase angle are directly optimized. Duty cycle of the DC-DC converter model: For each Buck converter and each Boost converter, its duty cycle decision variable is defined: Buck converter: (Real numbers, typically ranging from 0 to 1); Boost converter: (Real numbers, typically ranging from 0 to 1). Combine all decision variables into a vector. This serves as the decision space for the optimization model. The dimension of this decision space depends on the number of electric vehicle clusters, the number of VSCs, the number of DC / DC converters, and the number of optimization time slots. By setting the above decision variables, the optimization model can simultaneously control the charging and discharging behavior of electric vehicles, the voltage conversion capability of VSCs, and the voltage regulation capability of DC converters, thereby achieving coordinated optimization of the three-phase AC / DC distribution network.

[0049] Step S402: Input the initial battery state of charge value, target battery state of charge value, and power flow data of the three-phase AC / DC distribution network of each electric vehicle cluster into the optimization model. With minimizing the network loss cost function, the electric vehicle charging and discharging cost function, and the voltage deviation function as the optimization objective function, under the preset set of constraints, solve the optimization model through the preset optimization solution algorithm to obtain the optimal solution of each decision variable. In some embodiments, the following data is first input into the optimization model: the initial state of charge of each electric vehicle cluster. Target state of charge Maximum acceptable depth of discharge Battery capacity Charging efficiency Discharge efficiency Self-discharge rate Power flow data for the three-phase AC / DC distribution network (calculated by steps S301-S303), including: initial values ​​of three-phase and neutral point voltages at each AC node, initial voltages at each DC node, load power at each node (including ZIP loads), line parameters, VSC parameters (capacity, equivalent admittance, control mode), and DC / DC converter parameters (conductance, efficiency), etc.; time-of-use electricity price data. (yuan / kWh), and optimized time slot set (For example, dividing 24 hours into 24 cycles, each cycle being 1 hour, and each cycle being further divided into several time slots, with the time slot length being...) (Take 15 minutes). Next, establish the optimization objective function. The objective is to minimize the network loss cost function. Electric vehicle charging and discharging cost function Z EV and voltage deviation function Weighted sum: ; ; ; ; In the formula, This is the weighting coefficient (set to 1 or 0 according to actual operational needs). Weighting coefficients (set according to actual operational needs, for example, take...) ), For time slots The total active power loss of the entire network is obtained from power flow calculations; These are the sets of AC nodes and the sets of DC nodes, respectively. Indicates three phases; subscript The superscript indicates the real and imaginary parts of the voltage. The neutral point is indicated. Then, a set of constraints is set. These constraints include: AC / DC power flow equations and converter equations: including three-phase AC power flow equations, DC power flow equations, VSC coupling equations, Buck converter equations, and Boost converter equations (all calculated in the preceding steps). Equivalent electric vehicle equations: including the cluster SOC reaching the target value at the end of charging. At any given time, the SOC does not exceed the maximum. and not lower than the minimum : , Maximum discharge depth constraint: Charge-discharge mutual exclusion constraint: ;in, , They are nodes Cluster In the time slot The charging power and discharging power; These are the reciprocals of the charging efficiency and the discharging efficiency, respectively. Self-discharge rate; For model Battery capacity; The initial SOC and target SOC of the cluster; For model SOC upper and lower limits; The maximum acceptable depth of discharge; For calculating time slots. AC / DC node voltage upper and lower limit constraints: The upper and lower limits of the three-phase voltage at AC nodes are typically set to [0.9, 1.1] pu, and the upper limit of the neutral point voltage is set to 0.2 pu; the upper and lower limits of the DC node voltage are set to [0.9, 1.1] pu. AC line current upper limit constraint: DC line current upper limit constraint: Equation for calculating AC node injection current: And similar equations for DC node injection current. Finally, the optimization model is solved using a preset optimization algorithm. Since this optimization model is a nonlinear programming problem (the objective function is nonlinear, and the constraints include equality and inequality), gradient-based solvers, such as the CONOPT4 solver or IPOPT solver in GAMS software, or algorithms such as the interior-point method or Sequential Quadratic Programming (SQP), can be used. Algorithm parameters are set as follows: maximum number of iterations is 5000, and convergence accuracy is... For large-scale problems, the optimization period can be subdivided into multiple cycles (e.g., 24 cycles, each 1 hour). Optimization occurs independently within each cycle, with the result of the previous cycle used as the initial value for the next cycle, achieving rolling optimization. In each iteration, the solver calculates the objective function value and constraint violation rate, updating the decision variables until the convergence condition is met (i.e., the change in the objective function between two adjacent iterations is less than a preset threshold and all constraints are satisfied). After solving, the optimal solutions for each decision variable are obtained, including the optimal charging and discharging power for each electric vehicle cluster. The optimal voltage conversion ratio phasor for each VSC And the optimal duty cycle of each DC / DC converter. If the optimization fails to converge (e.g., the convergence condition is not met even after the number of iterations exceeds the upper limit), the voltage limit conditions are relaxed (e.g., the upper and lower limits of the AC node voltage are relaxed to [0.85, 1.15] pu, and the upper limit of the neutral point voltage is relaxed to 0.25 pu), and the optimization is repeated until convergence is achieved.

[0050] Step S403: Extract the target charging and discharging power of each electric vehicle cluster in the optimal solution.

[0051] In some embodiments, after the optimization solution in step S402 is completed, all decision variables related to the charging and discharging power of the electric vehicle cluster are extracted from the obtained optimal solution vector. Specifically: for each electric vehicle cluster (belongs to vehicle model) , connected to node ), in each optimization time slot Internally, extract: optimal charging power (If this value is 0, it means that the cluster is in the time slot) (No charging); Optimal discharge power (If this value is 0, it means that the cluster is in the time slot) (No discharge). Due to the mutual exclusion constraint between charging and discharging, the charging power and discharging power of the same cluster in the same time slot will not be positive simultaneously. The extracted optimal charging and discharging powers are arranged in chronological order to form the next day's charging and discharging power plan table for each electric vehicle cluster. The time resolution of this plan table is [missing information]. (For example, 15 minutes), covering the entire optimization period (e.g., 00:00 to 24:00). Finally, the target charging and discharging power commands for each electric vehicle cluster are sent to the corresponding charging piles or aggregator systems to control the actual vehicles to charge and discharge at the optimal power in the corresponding time slots. At the same time, the optimal voltage conversion ratio phasor of the VSC and the optimal duty cycle of the DC / DC converter are sent to the corresponding converter and converter controller to perform voltage regulation.

[0052] In some embodiments, the process of constructing the optimization objective function includes: obtaining the active power loss value and electricity price at each discrete time point within the charging and discharging time period of each electric vehicle, and summing the products of each active power loss value and each electricity price to obtain a network loss cost function; at each discrete time point, obtaining the node three-phase voltage, neutral point voltage, and node voltage of each AC node in the three-phase AC / DC distribution network, and calculating the deviations of each node three-phase voltage, each neutral point voltage, and each node voltage from the preset rated voltage, summing all the deviations to obtain a voltage deviation function; and weighting and summing the network loss cost function, the preset electric vehicle charging and discharging cost function, the voltage deviation function, and the preset weighting coefficients to obtain the optimization objective function.

[0053] In some embodiments, the active power loss value and electricity price at each discrete time point within the charging and discharging time period of each electric vehicle are obtained, and the products of each active power loss value and each electricity price are summed to obtain the network loss cost function. Specifically, let the optimized time slot set be... Each time slot The length is (Unit: hours). For each time slot The active power loss of the entire network is obtained through AC / DC power flow calculation. (Unit: kW), and obtain the electricity price corresponding to that time slot from the time-of-use electricity price data. (Unit: Yuan / kWh). Network Loss Cost Function Defined as the sum of the products of electricity loss and electricity price across all time slots: ; in, For network loss cost function (unit: yuan); To optimize the time slot set, for example, divide 24 hours into 96 time slots (one time slot every 15 minutes); For time slot index; For time slots The electricity price (yuan / kWh) varies depending on the peak, average, and off-peak periods. For time slots The total active power loss (kW) of the entire network is calculated from the AC / DC power flow, including AC line loss, DC line loss, VSC loss, and DC / DC converter loss. The time slot length (in hours), for example, 0.25 hours.

[0054] In some embodiments, at each of the discrete time points, the node three-phase voltage, neutral point voltage, and node voltage of each AC node in the three-phase AC / DC distribution network are obtained, and the deviations of each node three-phase voltage, each neutral point voltage, and each node voltage from the preset rated voltage are calculated respectively. All deviations are accumulated to obtain the voltage deviation function, specifically: for each time slot... Extract from the power flow calculation results: each communication node Each phase Real part of voltage and the virtual part (per-unit value), and the real part of the neutral point voltage. and the virtual part (per unit value); each DC node voltage amplitude (Per-unit value). The voltage deviation function is defined as the sum of three deviations: ; in, This is a voltage deviation function (dimensionless). For the set of communication nodes; For DC node set; To distinguish between phases, we take phases a, b, and c; , For communication nodes of The sum of the squares of the real and imaginary parts (per unit value) of the phase voltage is the square of the voltage amplitude; , For communication nodes The real and imaginary parts (per unit value) of the neutral point voltage; the rated neutral point voltage is 0. DC node The voltage amplitude (per unit) is 1.0 pu; this function suppresses voltage overshoot and neutral point voltage rise by minimizing the deviation of each voltage from the rated value.

[0055] It should be noted that the electric vehicle charging and discharging cost function for: ; in, For time slots Electricity price For time slots Discharge electricity price For car clusters Quantity, The total cost of charging and discharging an electric vehicle. A collection of all car models. For model The set of all clusters, To optimize the time slot set, For clusters In the time slot The charging power, For clusters In the time slot The discharge power.

[0056] In some embodiments, the network loss cost function, the preset electric vehicle charging and discharging cost function, the voltage deviation function, and the preset weighting coefficients are weighted and summed to obtain the optimization objective function. Specifically, the network loss cost function is... Electric vehicle charging and discharging cost function and voltage deviation function Through weighting coefficients By performing weighted summation, we obtain the comprehensive optimization objective function. : ; in, To optimize the objective function (unit and Consistency, is the element; when When dimensionless, Dimensions through Implicit conversion to meta); For network loss cost function (yuan); This is a voltage deviation function (dimensionless). and This is the weighting coefficient (a real number, set according to actual operational requirements). When =1 and At that time, only network loss costs and electric vehicle charging and discharging costs are optimized; when =0 and At that time, only network loss costs are optimized; when Taking larger values ​​(e.g., 10) places more emphasis on voltage quality optimization. Typical values ​​range from 0.1 to 10. This is achieved by minimizing... The optimized model effectively reduces the active power loss cost of the power grid and the charging and discharging cost of electric vehicles, while suppressing the deviation of three-phase voltage and neutral point voltage, achieving a coordinated optimization of economy and safety. The objective function, together with the aforementioned constraints (AC / DC power flow equations, equivalent electric vehicle equations, voltage and current limits, etc.), constitutes a complete optimization model, which is solved using a nonlinear programming solver (such as CONOPT4) to obtain the optimal values ​​of each decision variable.

[0057] In some embodiments, the set of constraints includes: AC / DC power flow constraints, upper and lower limits of electric vehicle state of charge constraints, mutual exclusion constraints of electric vehicle charging and discharging, maximum depth of discharge constraints of electric vehicles, upper and lower limits of AC / DC node voltage constraints, upper limit constraints of AC line current, upper limit constraints of DC line current, and node injection current constraints.

[0058] In some embodiments, the AC / DC power flow constraints, electric vehicle state of charge upper and lower limit constraints, electric vehicle charge and discharge mutual exclusion constraints, electric vehicle maximum discharge depth constraints, AC / DC node voltage upper and lower limit constraints, AC line current upper limit constraints, DC line current upper limit constraints, and node injection current constraints are specifically as follows: (1) AC / DC power flow constraints and converter equations: including three-phase AC power flow equations, DC power flow equations, VSC coupling equations, Buck converter equations, and Boost converter equations (all of which have been given calculation expressions in the preceding steps). (2) Equivalent electric vehicle equations: including clusters at the end of charging. The SOC has reached the target value: At any given time, the SOC does not exceed the maximum. and not lower than the minimum : , Maximum discharge depth constraint: Charge-discharge mutual exclusion constraint: ;in, , They are nodes Cluster In the time slot The charging power and discharging power; These are the reciprocals of the charging efficiency and the discharging efficiency, respectively. Self-discharge rate; For model Battery capacity; The initial SOC and target SOC of the cluster; For model SOC upper and lower limits; The maximum acceptable depth of discharge; For calculating time slots. (3) Upper and lower limits of AC / DC node voltage constraints: Among them, the upper and lower limits of the three-phase voltage at the AC node are usually set to [0.9, 1.1] pu, and the upper limit of the neutral point voltage is set to 0.2 pu; the upper and lower limits of the DC node voltage are set to [0.9, 1.1] pu. (4) Upper limit constraint of AC line current: ,in, , For time slots The real and imaginary parts (A) of the current. (5) Maximum line current (A) constraint: ,in, For time slots DC lines The current (A). Its upper limit (A). (6) Equation for calculating AC node injection current: ,in, Injecting complex current vectors into constant impedance, constant current, and constant power loads other than electric vehicles. A complex vector of current injected into an electric vehicle; and similar equations for the current injected into a DC node: The above constraints together constitute the feasible region of the optimization model, ensuring that the optimization results meet the requirements of safe grid operation and electric vehicle charging.

[0059] Through the above steps, the power flow distribution of the three-phase AC / DC distribution network was accurately solved, and an optimization model was constructed with the goal of minimizing network loss costs and voltage deviation. The charging and discharging power of electric vehicle clusters, VSC voltage conversion ratio, and DC / DC converter duty cycle were coordinated and optimized. Under the premise of meeting the grid safety constraints and electric vehicle charging needs, the network loss costs and voltage over-limit risks were effectively reduced, thereby improving the economy and reliability of large-scale electric vehicle charging and discharging control in three-phase AC / DC distribution networks.

[0060] Step S104: Perform charge and discharge control on each of the electric vehicles based on the target charge and discharge power.

[0061] In some embodiments, firstly, each electric vehicle cluster is extracted from the optimal solution of the optimization model. In each time slot Target charge and discharge power and For each cluster This cluster consists of multiple electric vehicles of the same type, with similar parking and charging times, and similar initial and target states of charge (SOC). Its equivalent charge / discharge power represents the total charge / discharge power of all vehicles within the cluster. Then, based on the cluster... The actual number of vehicles included The total target power of the cluster is proportionally allocated to each electric vehicle. The allocation rules are as follows: If Then the charging power command for each electric vehicle in the cluster is: The discharge power command is 0; if Then the discharge power command for each electric vehicle in the cluster is: If the charging power command is 0, all electric vehicles in the cluster will neither charge nor discharge. For vehicles with different battery capacities or rated power within the cluster (e.g., batteries of the same model but with different aging levels), a weighted allocation based on battery capacity ratio can be performed to ensure consistent charge / discharge rates for all vehicles. Next, the charging / discharge power command for each electric vehicle is sent to the corresponding charging equipment via the aggregator unit (AU) or directly through the charging pile's communication interface. This sending occurs before the start of each optimization cycle (e.g., within 5 minutes before the start of each hour) to ensure the charging pile performs charging and discharging according to the commanded power in the next cycle (e.g., 1 hour). The charging pile controller switches the charging / discharging power according to the command in the corresponding time slot. During execution, if an electric vehicle leaves early (actual parking time shorter than expected), the system records its actual charging amount and redistributes the expected SOC of the remaining vehicles in its cluster in subsequent cycles to ensure that the remaining vehicles still meet their charging targets. If a vehicle's actual parking time is longer than expected, it continues to execute the last issued command until its actual departure time, without additional charging. Furthermore, if the optimization calculation for a certain cycle fails to converge, the voltage limit conditions are relaxed (e.g., the upper and lower limits of the AC node voltage are relaxed from [0.9, 1.1] pu to [0.85, 1.15] pu, and the upper limit of the neutral point voltage is relaxed from 0.2 pu to 0.25 pu) and the optimization is repeated until a convergent optimal solution is obtained. The charging and discharging power corresponding to this convergent solution is issued as the control command for that cycle. For vehicle clusters that stop before 24:00, their target SOC has been evenly distributed across cycles according to the proportion of the time before 24:00 to the total parking time. Therefore, the optimized charging and discharging power command has already taken into account the charging demand across days. During execution, after 24:00 (i.e., at midnight the next day), control continues according to the optimization results of the new cycle until the vehicles actually leave. Finally, after all cycles have been executed on the next day, the system records the actual charging and discharging capacity, actual SOC changes, and grid operation data for each electric vehicle to update the user's historical behavior database and provide data support for subsequent optimizations.

[0062] Through the above steps, the target charging and discharging power obtained from the optimization solution is proportionally allocated to each electric vehicle and precisely executed through the aggregator system or the charging pile communication interface. At the same time, corresponding processing mechanisms are designed for actual scenarios such as vehicles leaving early, extending parking time, optimization non-convergence, and charging across days, ensuring the reliable implementation and dynamic adaptability of control commands. This enables efficient and robust control of large-scale electric vehicle charging and discharging in a three-phase AC / DC power distribution network.

[0063] like Figure 5 As shown, based on the above method embodiments, corresponding apparatus embodiments are provided; An embodiment of the present invention provides a schematic diagram of the structure of a charging and discharging control system for an electric vehicle, including: an acquisition module 100, a clustering module 200, a solution module 300, and a control module 400; The acquisition module 100 is used to acquire charging and discharging time period data and battery status data of several electric vehicles, as well as topology parameters and power load data of the three-phase AC / DC distribution network. The clustering module 200 is used to cluster each electric vehicle based on the charging and discharging time period data and the battery state data to obtain several electric vehicle clusters. The solution module 300 is used to input the topology parameters, the power load data, and the initial charging and discharging power of each electric vehicle into a preset power flow equation to obtain the power flow data of the three-phase AC / DC distribution network. The power flow data, the initial battery state of charge value and the target battery state of charge value of each electric vehicle cluster are input into a preset optimization model. The optimization model is solved with the goal of minimizing the network loss cost function, the electric vehicle charging and discharging cost function and the voltage deviation function to obtain the target charging and discharging power of each electric vehicle cluster. The control module 400 is used to control the charging and discharging of each of the electric vehicles based on the target charging and discharging power.

[0064] It is understood that the above-described device embodiments correspond to the method embodiments of the present invention, and can implement the charging and discharging control method for an electric vehicle provided by any of the above-described method embodiments of the present invention. More detailed workflows and principles of this system can be found, but are not limited to, the relevant descriptions of the above methods.

[0065] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can specifically be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0066] Based on the above embodiment of the charging and discharging control method for an electric vehicle, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the charging and discharging control method for an electric vehicle according to any embodiment of the present invention.

[0067] For example, in this embodiment, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the terminal device.

[0068] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0069] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0070] Based on the above-described method embodiments, another embodiment of the present invention provides a computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the charging and discharging control method for an electric vehicle as described in any of the above-described method embodiments of the present invention.

[0071] The modules / units integrated in the device / terminal equipment, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0072] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A charging and discharging control method for an electric vehicle, characterized in that, include: Acquire data on the charging and discharging time periods and battery status of several electric vehicles, as well as the topology parameters and power load data of the three-phase AC / DC distribution network; Based on the charging and discharging time period data and the battery state data, the electric vehicles are clustered to obtain several electric vehicle clusters; The topology parameters, the power load data, and the initial charging and discharging power of each electric vehicle are input into a preset power flow equation to obtain the power flow data of the three-phase AC / DC distribution network. The power flow data, the initial battery state of charge value and the target battery state of charge value of each electric vehicle cluster are input into a preset optimization model. The optimization model is solved with the goal of minimizing the network loss cost function, the electric vehicle charging and discharging cost function and the voltage deviation function to obtain the target charging and discharging power of each electric vehicle cluster. The electric vehicles are charged and discharged based on the target charging and discharging power.

2. The charging and discharging control method for an electric vehicle as described in claim 1, characterized in that, The electric vehicles are clustered based on the charging / discharging time period data and the battery state data to obtain several electric vehicle clusters, including: Extract the parking charging time of each electric vehicle from the charging and discharging time period data, and extract the initial state of charge value and target state of charge value of each electric vehicle from the battery state data. Based on the parking charging time, the initial state of charge value, and the target state of charge value, the electric vehicles are clustered to merge electric vehicles of the same type that have similar parking charging times, the same initial state of charge value and target state of charge value, and similar acceptable depth of discharge, thus obtaining several electric vehicle clusters.

3. The charging and discharging control method for an electric vehicle as described in claim 1, characterized in that, The process of inputting the topology parameters, the power load data, and the initial charging and discharging power of each electric vehicle into a preset power flow equation to obtain the power flow data of the three-phase AC / DC distribution network includes: The initial injection current and initial node voltage of each node in the three-phase AC / DC distribution network are determined based on the topology parameters, the power load data, and the initial charging and discharging power. Substitute each of the initial injection currents and each of the initial node voltages into the power flow equations, and iteratively solve the power flow equations using a preset Newton iteration method. In each iteration, calculate the current imbalance of each node, and correct each of the initial node voltages based on the current imbalance to obtain the target node voltage. The injected current of each node in the three-phase AC / DC distribution network is calculated based on the target node voltage, and it is determined whether the node current imbalance of each injected current meets the preset convergence condition. If it meets the condition, the iteration stops, and the power flow data of the three-phase AC / DC distribution network is output. The power flow data includes the voltage, current and power distribution results of each node.

4. The charging and discharging control method for an electric vehicle as described in claim 3, characterized in that, The process of constructing the power flow equations includes: Based on the topology parameters, construct the node admittance matrix of the three-phase AC power grid and the node admittance matrix of the DC power grid, and determine the power flow equations of the three-phase AC power grid and the DC power grid respectively based on the node admittance matrix of the three-phase AC power grid and the node admittance matrix of the DC power grid. By combining the preset three-phase voltage source converter model, the preset DC converter model, the three-phase AC power flow equation, and the DC power flow equation, a power flow equation including the three-phase AC power grid, the DC power grid, and the AC / DC conversion device is obtained.

5. The charging and discharging control method for an electric vehicle as described in claim 4, characterized in that, The optimization model is solved by minimizing the network loss cost function, the electric vehicle charging and discharging cost function, and the voltage deviation function as the optimization objective functions to obtain the target charging and discharging power of each electric vehicle cluster, including: Within the decision space of the optimization model, the charging and discharging power of each electric vehicle cluster, the voltage conversion ratio phasor of the three-phase voltage source converter model, and the duty cycle of the DC-DC converter model are determined as decision variables. The initial battery state of charge (SOC) value, target battery SOC value, and power flow data of the three-phase AC / DC distribution network of each electric vehicle cluster are input into the optimization model. The optimization objective function is to minimize the network loss cost function, the electric vehicle charging and discharging cost function, and the voltage deviation function. Under the preset set of constraints, the optimization model is solved by the preset optimization solution algorithm to obtain the optimal solution of each decision variable. Extract the target charge / discharge power of each electric vehicle cluster from the optimal solution.

6. The charging and discharging control method for an electric vehicle as described in claim 5, characterized in that, The process of constructing the optimization objective function includes: The active power loss value and electricity price at each discrete time point within the charging and discharging time period of each electric vehicle are obtained, and the products of each active power loss value and each electricity price are accumulated and summed to obtain the network loss cost function. At each of the discrete time points, the node three-phase voltage, neutral point voltage and node voltage of each AC node in the three-phase AC / DC distribution network are obtained, and the deviations of each node three-phase voltage, each neutral point voltage and each node voltage from the preset rated voltage are calculated respectively. All the deviations are accumulated to obtain the voltage deviation function. The network loss cost function, the preset electric vehicle charging and discharging cost function, the voltage deviation function, and the preset weighting coefficients are weighted and summed to obtain the optimization objective function.

7. The charging and discharging control method for an electric vehicle as described in claim 5, characterized in that, The set of constraints includes: AC / DC power flow constraints, upper and lower limits of electric vehicle state of charge constraints, mutual exclusion constraints of electric vehicle charging and discharging constraints, maximum depth of discharge constraints of electric vehicles, upper and lower limits of AC / DC node voltage constraints, upper limit constraints of AC line current, upper limit constraints of DC line current, and node injection current constraints.

8. A charging and discharging control system for an electric vehicle, characterized in that, The system includes: an acquisition module, a clustering module, a solution module, and a control module; The acquisition module is used to acquire charging and discharging time period data and battery status data of several electric vehicles, as well as topology parameters and power load data of the three-phase AC / DC distribution network. The clustering module is used to cluster each electric vehicle based on the charging and discharging time period data and the battery state data to obtain several electric vehicle clusters. The solution module is used to input the topology parameters, the power load data, and the initial charging and discharging power of each electric vehicle into a preset power flow equation to obtain the power flow data of the three-phase AC / DC distribution network. The power flow data, the initial battery state of charge value and the target battery state of charge value of each electric vehicle cluster are input into a preset optimization model. The optimization model is solved with the goal of minimizing the network loss cost function, the electric vehicle charging and discharging cost function and the voltage deviation function to obtain the target charging and discharging power of each electric vehicle cluster. The control module is used to control the charging and discharging of each of the electric vehicles based on the target charging and discharging power.

9. A terminal device, characterized in that, The device includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements a charging and discharging control method for an electric vehicle as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, include: A stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform a charging and discharging control method for an electric vehicle as described in any one of claims 1-7.

Images