Management method for accessing electric vehicle to power distribution network

Abstract

The invention discloses a management method for accessing an electric vehicle to a power distribution network. The method comprises the following steps: S1, a vehicle-network interaction mechanism; s2, dynamically optimizing charging and discharging power; s3, performing active-reactive collaborative optimization; s4, communication and data architecture; s5, adaptive weight adjustment is carried out; s6, verification and evaluation; the historical data comprises a charging period, an electric quantity demand and cost sensitivity, the user types comprise a rigid charging type, a flexible charging and discharging type and an economic priority type, and the simulation test platform adopts MATLAB software to call OpenDSS and is used for establishing, compiling and solving a circuit model. According to the invention, through the hierarchical optimization architecture and the multi-time scale control strategy, the cooperation of the demand preference of the vehicle owner, the dynamic regulation and control of the charging and discharging power and the economic operation target of the power grid is realized, so that the power distribution network accessed by the high-permeability electric vehicle is more standardized.

Description

Technical Field

[0001] The present invention relates to the technical field of electric vehicles, and specifically to a management method for electric vehicles accessing the distribution network. Background Art

[0002] The large-scale access of electric vehicles will increase the power demand. Especially during peak hours, it may cause the distribution network to be overloaded, affecting the power supply reliability and voltage stability. At this time, it is necessary to manage the charging time and power to avoid excessive grid pressure. As a distributed energy storage unit, if properly managed, electric vehicles can provide power support when the grid needs it. For example, charging during low electricity consumption periods and discharging during peak periods, thus balancing the grid load, reducing the dependence on traditional power generation facilities, and improving energy utilization efficiency. In addition, the needs of users are diverse. For example, some users hope to charge their vehicles as quickly as possible, while others are more concerned about the charging cost. Through management, the charging strategy can be optimized to meet the needs of different users while reducing the overall charging cost.

[0003] Managing the access of electric vehicles to the distribution network can ensure grid stability, improve energy efficiency, meet user needs, address technical challenges, and make full use of the flexible resources of electric vehicles to achieve the economic, safe, and efficient operation of the grid.

[0004] Therefore, it is necessary to manage the access of electric vehicles to the distribution network. Based on this, we propose to design a management method for electric vehicles accessing the distribution network. Summary of the Invention

[0005] The purpose of this part is to outline some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this part, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this part, the abstract, and the title. However, such simplifications or omissions cannot be used to limit the scope of the present invention.

[0006] To solve the above technical problems, according to one aspect of the present invention, the following technical solutions are provided:

[0007] A management method for electric vehicles accessing the distribution network, comprising:

[0008] S1. Vehicle-grid interaction mechanism

[0009] S1.1 Demand preference modeling

[0010] User portrait construction: Based on historical data, establish a database of vehicle owner demand preferences and classify user types;

[0011] Dynamic behavior capture: Introduce a Markov decision process to describe the dynamic switching logic of the vehicle owner's charging mode. Combine with a real-time interaction interface and the preference settings feedback of users on the APP to update the user status;

[0012] S1.2 Energy Boundary Model

[0013] Time - series Energy Constraint:

[0014] Establish the upper and lower limit models of charging and discharging power for an electric vehicle cluster:

[0015]

[0016] Among them, the power boundary is jointly determined by the battery SOC, the capacity of the charging pile, and the user - preset cut - off time;

[0017] S2. Charging and Discharging Power Dynamic Optimization

[0018] S2.1 Hierarchical Control Architecture

[0019] Distribution network layer: Based on model predictive control, roll - optimizes the total power demand of the distribution network, with the goals of minimizing network loss and voltage deviation, and issues the cluster power command to the regional controller;

[0020] Electric vehicle cluster layer: Adopts a distributed optimization algorithm to allocate individual charging and discharging power under the constraints of meeting the owner's preferences, and minimizes the deviation between the total power and the command;

[0021] S2.2 Real - time Correction Strategy

[0022] Dynamic Power Redistribution:

[0023] Design an event - trigger mechanism. When the owner actively modifies the charging mode or the grid state changes suddenly, trigger local re - optimization to ensure the real - time balance between user needs and grid security;

[0024] S3. Active - Reactive Power Co - optimization

[0025] S3.1 Charging Pile VSC Control Model

[0026] Establish the active - reactive decoupling control equation of the intelligent charging pile converter:

[0027] P = V g I d , Q = - V g I q

[0028] Analyze the constraint relationship between active power demand and reactive power regulation ability;

[0029] S3.2 Multi - objective Co - optimization

[0030] Mixed - integer Second - order Cone Programming Model:

[0031] With the dual objectives of minimizing the operation cost of the distribution network and minimizing the voltage deviation, construct an optimization problem:

[0032] min(α∑C grid,t +β∑|V i,t -V nom |)

[0033] The constraints include power flow equations, the operating range of electric vehicles, and the output limits of distributed power sources;

[0034] S4. Communication and Data Architecture

[0035] S4.1 Edge Computing Node Deployment: Deploy edge servers on the distribution transformer side to facilitate millisecond-level local power distribution calculation and reduce cloud communication latency;

[0036] S4.2 Blockchain Evidence Storage: Use a lightweight blockchain to record the interaction logs of user preferences and grid commands to facilitate ensuring data credibility;

[0037] S5. Adaptive Weight Adjustment

[0038] Fuzzy Logic Controller: Dynamically adjust the optimization target weights according to the real-time load rate of the distribution network. Among them, voltage stability is prioritized during heavy loads, and economy is emphasized during light loads;

[0039] S6. Verification and Evaluation

[0040] S6.1 Simulation Test Platform

[0041] Joint Simulation Environment: Build an active distribution network - electric vehicle joint simulation platform through MATLAB / OpenDSS to verify the voltage fluctuation suppression effect and economic benefits;

[0042] S6.2 Actual Effect Indicators

[0043] Define the demand deviation degree index

[0044] As a preferred solution of a management method for electric vehicles to access the distribution network according to the present invention, in S1, the historical data includes charging time periods, electricity demand, and cost sensitivity, and the user types include rigid charging type, flexible charge and discharge type, and economy - priority type.

[0045] As a preferred solution of a management method for electric vehicles to access the distribution network according to the present invention, in S6, the simulation test platform uses MATLAB software to call OpenDSS for establishing, compiling, and solving the circuit model, enabling users to utilize the powerful data processing and visualization functions of MATLAB, and at the same time using the professional power grid simulation capabilities of OpenDSS to achieve efficient power grid analysis and design.

[0046] As a preferred solution of a management method for electric vehicles accessing the distribution network according to the present invention, in S6, the deviation index D < 5%.

[0047] As a preferred solution of a management method for electric vehicles accessing the distribution network according to the present invention, in S1, the algorithm for classifying user types is the K-means algorithm. When processing large data sets such as user types, the K-means algorithm is usually more efficient than other clustering algorithms.

[0048] Compared with the prior art, the beneficial effects of the present invention are:

[0049] Through the hierarchical optimization architecture and multi-time scale control strategy, the present invention realizes the coordination of the owner's demand preferences, dynamic regulation of charging and discharging power, and the goal of grid economic operation, thus facilitating the standardization of the distribution network with high-penetration electric vehicles. Detailed implementation manners

[0050] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following further describes the implementation manners of the present invention in detail.

[0051] The present invention provides a management method for electric vehicles accessing the distribution network, including:

[0052] S1. Vehicle-grid interaction mechanism

[0053] S1.1 Demand preference modeling

[0054] User portrait construction: Based on historical data, a database of owner demand preferences is established to classify user types; among them, historical data includes charging time periods, power demand, and cost sensitivity, and user types include rigid charging type, flexible charging and discharging type, and economy priority type. The algorithm for classifying user types is the K-means algorithm. When processing large data sets such as user types, the K-means algorithm is usually more efficient than other clustering algorithms.

[0055] Dynamic behavior capture: Introduce a Markov decision process to describe the dynamic switching logic of the owner's charging mode, and combine it with a real-time interaction interface and the user's preference settings feedback on the APP to update the user status;

[0056] S1.2 Energy boundary model

[0057] Time-series energy constraint:

[0058] Establish an upper and lower limit model for the charging and discharging power of an electric vehicle cluster:

[0059]

[0060] P EV,t:Total active power of the electric vehicle cluster at time t. A positive value indicates power absorption from the grid, i.e., charging, and a negative value indicates power injection into the grid, i.e., discharging.

[0061] Q EV,t :Total reactive power of the electric vehicle cluster at time t. A positive value indicates inductive reactive power compensation, and a negative value indicates capacitive reactive power compensation.

[0062] The upper and lower limits of the active power are determined by the following factors:

[0063] (1) Battery charge and discharge rate limit;

[0064] (2) Rated power of the charging pile;

[0065] (3) The charging cut-off time set by the user, which needs to ensure full charge before the cut-off time.

[0066] The upper and lower limits of the reactive power are determined by the capacity of the charging pile converter and the grid voltage support requirement.

[0067] Through the above model, it is convenient to define the power regulation ability boundary of the electric vehicle cluster at each moment, ensuring that the charging and discharging behavior meets the user's needs and equipment safety.

[0068] S2. Dynamic Optimization of Charging and Discharging Power

[0069] S2.1 Hierarchical Control Architecture

[0070] Distribution network layer: Based on model predictive control, the total power demand of the distribution network is optimized in a rolling manner, aiming at minimizing network loss and voltage deviation, and issuing the cluster power command to the regional controller;

[0071] Electric vehicle cluster layer: Adopt a distributed optimization algorithm to allocate individual charging and discharging power under the constraint of meeting the owner's preference, and minimize the deviation between the total power and the command;

[0072] S2.2 Real-time Correction Strategy

[0073] Dynamic Power Redistribution:

[0074] Design an event-triggered mechanism. When the owner actively modifies the charging mode or the grid state changes suddenly, trigger local re-optimization to ensure real-time balance between user needs and grid safety;

[0075] S3. Active-Reactive Coordinated Optimization

[0076] S3.1 Charging Pile VSC Control Model

[0077] Establish the active-reactive decoupling control equation of the intelligent charging pile converter:

[0078] P = V g I d , Q = -V g I q

[0079] V g : Grid voltage, referring to the voltage amplitude at the converter connection point;

[0080] I d : Direct axis of current, i.e., d-axis component, which is directly related to active power;

[0081] I q : Quadrature axis of current, i.e., q-axis component, which is directly related to reactive power.

[0082] By adjusting I d and I q , the active power P and reactive power Q can be independently controlled. When the electric vehicle charges at a constant power, P is fixed, and at this time, the adjustment ability of Q is limited by the total capacity S max of the converter, and the formula needs to be satisfied:

[0083] S3.2 Multi-objective collaborative optimization

[0084] Mixed-integer second-order cone programming model:

[0085] Taking the minimum operation cost of the distribution network and the minimum voltage deviation as two objectives, an optimization problem is constructed:

[0086] min(α∑C grid,t +β∑|V i,t -V nom |)

[0087] C grid,t : Grid power purchase cost at time t, which is related to the time-of-use electricity price;

[0088] V i,t : Voltage amplitude of node i in the distribution network at time t;

[0089] V nom : Nominal value of the node voltage;

[0090] α, β: Weight coefficients, used to balance economy and voltage stability; for example, when the distribution network is heavily loaded, then set β > α to give priority to ensuring voltage stability;

[0091] By optimizing the charging and discharging power and reactive power output of electric vehicles, the trade-off between the economy and security of the distribution network is achieved.

[0092] S4. Communication and data architecture

[0093] S4.1 Edge Computing Node Deployment: Deploy edge servers on the side of distribution transformers to facilitate millisecond-level local power distribution calculation and reduce cloud communication latency;

[0094] S4.2 Blockchain Evidence Storage: Use lightweight blockchain to record the interaction logs of user preferences and grid instructions to facilitate ensuring data credibility;

[0095] S5. Adaptive Weight Adjustment

[0096] Fuzzy Logic Controller: Dynamically adjust the optimization target weights according to the real-time load rate of the distribution network. When it is heavily loaded, voltage stability is prioritized, and when it is lightly loaded, economy is emphasized;

[0097] S6. Verification and Evaluation

[0098] S6.1 Simulation Test Platform

[0099] Joint Simulation Environment: Build an active distribution network - electric vehicle joint simulation platform to verify the voltage fluctuation suppression effect. Under typical scenarios, the voltage deviation ≤ 2%, and there is an economic benefit with a 15% - 25% cost reduction.

[0100] In this embodiment, the simulation test platform uses MATLAB software to call OpenDSS for establishing, compiling, and solving the circuit model, enabling users to utilize the powerful data processing and visualization functions of MATLAB and, at the same time, the professional power grid simulation capabilities of OpenDSS to achieve efficient power grid analysis and design.

[0101] S6.2 Actual Effect Indicators

[0102] Define the demand deviation degree indicator The deviation degree indicator D < 5%.

[0103] SOC actual : The actual charging state of the electric vehicle battery, with a range of 0% - 100%;

[0104] SOC target : The target charging state set by the user;

[0105] N: The total number of electric vehicles counted;

[0106] D: Represents the average deviation degree of the user's charging demand. If D < 5%, it indicates that the actual charging result highly coincides with the user's expectation, ensuring user satisfaction.

[0107] Although the present invention has been described above with reference to the embodiments, various modifications can be made thereto and components thereof can be replaced with equivalents without departing from the scope of the present invention. In particular, as long as there is no structural conflict, the features in the embodiments disclosed in the present invention can be combined with each other in any manner, and the reason for not exhaustively describing the situations of these combinations in this specification is only for saving space and resources. Therefore, the present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A management method for an electric vehicle to access a distribution network, characterized in that, Including: S1, Vehicle-grid Interaction Mechanism S1.1 Requirement Preference Modeling User Portrait Construction: Establish a demand preference library for vehicle owners based on historical data and divide user types; Dynamic Behavior Capture: Introduce the Markov decision process to describe the dynamic switching logic of vehicle owners' charging modes. Combine with the real-time interaction interface and the preference settings feedback of users on the APP to update the user status; S1.2 Energy Boundary Model Time-series Energy Constraint: Establish the upper and lower limit models of the charging and discharging power of the electric vehicle cluster: S2, Dynamic Optimization of Charging and Discharging Power S2.1 Hierarchical Control Architecture Distribution Network Layer: Based on model predictive control, roll-optimize the total power demand of the distribution network, aiming at minimizing network loss and voltage deviation, and send the cluster power command to the regional controller; Electric Vehicle Cluster Layer: Adopt a distributed optimization algorithm to allocate the individual charging and discharging power under the constraint of meeting the vehicle owners' preferences, and minimize the deviation between the total power and the command; S2.2 Real-time Correction Strategy Dynamic Power Reallocation: Design an event-triggering mechanism. When the vehicle owner actively modifies the charging mode or the grid state changes suddenly, trigger local re-optimization to ensure the real-time balance between user needs and grid security; S3, Active-reactive Power Co-optimization S3.1 Charging Pile VSC Control Model Establish the active-reactive decoupling control equation of the intelligent charging pile converter: P = V g I d , Q = -V g I q Analyze the constraint relationship between the active power demand and the reactive power regulation ability; S3.2 Multi-objective Co-optimization Mixed Integer Second-order Cone Programming Model: Take the minimum operation cost of the distribution network and the minimum voltage deviation as the dual objectives to construct an optimization problem: min(α∑C grid,t +β∑|V i,t -V nom |) The constraints include power flow equations, the operating range of electric vehicles, and the output limits of distributed power sources; S4, Communication and Data Architecture S4.1 Edge Computing Node Deployment: Deploy edge servers on the side of the distribution transformer; S4.2 Blockchain Evidence Storage: Adopt a lightweight blockchain to record the interaction logs of user preferences and grid commands; S5, Adaptive Weight Adjustment Fuzzy Logic Controller: Dynamically adjust the optimization target weights according to the real-time load rate of the distribution network; S6, Verification and Evaluation S6.1 Simulation Test Platform Joint Simulation Environment: Build an active distribution network - electric vehicle joint simulation platform through MATLAB / OpenDSS to verify the voltage fluctuation suppression effect and economic benefits; S6.2 Actual Effect Index Define the requirement deviation degree index 2. The management method for an electric vehicle accessing a distribution network according to claim 1, characterized in that, In S1, the historical data includes charging time period, power demand, and cost sensitivity, and the user types include rigid charging type, flexible charging and discharging type, and economy priority type.

3. The management method for an electric vehicle to access a distribution network according to claim 1, characterized in that, In S6, the simulation test platform uses MATLAB software to call OpenDSS for the establishment, compilation, and solution of the circuit model.

4. A management method for an electric vehicle to access a distribution network according to claim 1, characterized in that, In S6, the deviation index D < 5%.

5. The management method for an electric vehicle to access a distribution network according to claim 1, characterized in that, In S1, the algorithm for dividing user types is the K-means algorithm.

Images