Micro-grid electric vehicle ordered charging scheduling method considering differential charging demands

Abstract

The invention provides a micro-grid electric vehicle orderly charging scheduling method considering differentiated charging requirements. The method comprises the following steps: acquiring basic parameters and charging behavior data of multiple types of electric vehicles, and constructing a charging behavior probability model and a schedulable capacity evaluation model; constructing an island microgrid master-slave game double-layer optimization model containing the electric vehicle based on the charging characteristics of the electric vehicle and the differentiated charging requirements of users; a model conversion and solution method based on a KKT condition is utilized to convert a double-layer optimization problem into a single-layer mixed integer linear programming problem, and a global optimal scheduling scheme is efficiently solved; finally, the time transfer and peak-valley difference minimization of the charging load of the electric vehicle are realized, the peak load pressure of the micro-grid is relieved, and the admitting ability of the power grid to large-scale electric vehicles is improved. According to the invention, power grid operation efficiency, safety and user charging individuation requirements are considered, and collaborative optimization of ordered charging of the electric vehicle and flexible and efficient operation of the micro-grid is realized.

Description

Technical Field

[0001] This application belongs to the field of electric vehicle scheduling technology, and in particular relates to a microgrid electric vehicle orderly charging scheduling method that takes into account differentiated charging needs. Background Technology

[0002] The widespread use of electric vehicles (EVs) is an important way to replace electricity, revolutionize energy consumption, and reduce environmental pollution. However, the concentrated charging of a large number of EVs at the same time will significantly widen the peak-valley difference in the power grid, reduce power quality, increase operating costs, and may jeopardize the safe and stable operation of microgrids. In fact, EV batteries have millisecond-level response capabilities, and some vehicles have an average daily idle time of more than 20 hours, possessing considerable dispatchable potential. Through reasonable charge and discharge control, EV clusters can be transformed into active energy storage units, thereby participating in demand response and achieving the dual functions of peak shaving and valley filling as well as renewable energy consumption.

[0003] Currently, EV charging faces the following pain points. First, the charging power exhibits a step-like jump, and efficiency drops rapidly after the State of Charge (SOC) reaches the threshold, making it difficult for traditional constant power models to depict the true characteristics. Second, large-scale passive access and disorderly charging of EVs cause negative effects such as local overload of the microgrid, voltage exceeding limits, and premature aging of transformers.

[0004] Therefore, it is also necessary to build a collaborative scheduling framework that takes into account the physical characteristics of EV charging, the differentiated needs of users, and the operational constraints of microgrids, so as to transform the charging load from a rigid demand into a flexible resource and minimize the peak-valley difference of microgrids while ensuring users' travel. Summary of the Invention

[0005] To overcome the problems of indiscriminate charging behavior, neglect of differences, rough estimation of schedulable capacity, and aggravated peak-valley differences in related technologies, this application provides a method and device for coordinated charging scheduling of electric vehicles under islanded microgrids that considers differentiated charging needs. It can achieve optimal microgrid operating costs and minimize peak-valley differences while ensuring users' personalized travel and charging needs, thus achieving multiple objectives such as orderly charging, peak shaving and valley filling, and high proportion of new energy consumption.

[0006] This application is achieved through the following technical solution: In a first aspect, embodiments of this application provide a method for coordinated charging scheduling of electric vehicles considering differentiated charging demands in an islanded microgrid, including: Obtain basic parameters for various types of electric vehicles; Based on fundamental parameters and the coupling relationship between SOC, power, and range, a three-state schedulable capacity evaluation model is constructed, which includes charge-only, discharge-only, and charge / discharge-only states. Monte Carlo simulation is used to generate minute-level schedulable capacity curves. The charging urgency index CUI is introduced to classify EVs into fast charging / slow charging modes, and a master-slave game two-layer optimization model including the microgrid layer and the electric vehicle scheduling layer is established. In the master-slave game two-level optimization model, the upper-level model aims to minimize the overall operating cost of the microgrid, while the lower-level model aims to minimize the peak-valley load difference. Based on KKT conditions, the Big-M method and the strong duality theorem, the master-slave game two-level optimization model is transformed into a single-level mixed integer linear programming problem, and a convex optimization solver is used to obtain the globally optimal scheduling scheme. The optimal solution is sent to the charging pile controller to achieve real-time and orderly control of EV charging time and power.

[0007] In one embodiment, the three-state schedulable capacity assessment model includes: Charging-only scheduling: SOC < SOCmin and already connected to a charging station; Discharge-only scheduling: SOC > SOCmax and has V2G capability; Both charging and discharging can be scheduled: SOCmin≤SOC≤SOCmax and the vehicle is online.

[0008] In one embodiment, the CUI metric is defined as: When CUIi < 0, it is determined to be emergency fast charging; when CUIi ≥ 0, it is determined to be slow charging.

[0009] In one embodiment, the master-slave game two-layer optimization model includes: The objective function of the upper microgrid layer is to minimize "unit fuel cost + energy storage depreciation cost + power outage penalty cost"; The objective function of the lower-level EV scheduling layer is to minimize the "peak-valley difference of the total microgrid load". The constraints include power balance, upper and lower limits of unit output, energy storage SOC, number of charging piles, and fast / slow charging power boundaries.

[0010] In one embodiment, the KKT conversion step includes: Construct the lower-level Lagrangian function → Introduce KKT and complementary relaxation conditions → Use Big-M linearization for complementary relaxation → Eliminate the upper-level objective nonlinear term using strong duality → Form a single-level mixed-integer linear programming problem → Solve using Gurobi / Cplex and verify optimality by backtracking KKT.

[0011] Secondly, embodiments of this application provide a coordinated charging scheduling device for electric vehicles considering differentiated charging needs under an islanded microgrid, applying the method described in the first aspect, including: The data acquisition module is used to collect EV battery capacity, SOC, access / departure time, charging pile status, and microgrid operating parameters; The schedulable capacity assessment module is used to generate three-state schedulable capacity curves. The master-slave game modeling module is used to construct a two-layer optimization model and set the objectives and constraints of the upper and lower layers; The KKT transformation and solution module is used to transform a two-level problem into a single-level mixed integer linear programming problem and output the global optimal solution. The scheduling and execution module is used to distribute the optimal charging plan to the charging piles to achieve orderly charging and peak shaving and valley filling.

[0012] The beneficial effects of the embodiments in this application compared with the prior art are: In the scenario of isolated microgrids, a collaborative scheduling framework is proposed that takes into account the three-state schedulable capacity of EVs, the urgency of user CUI, and master-slave game theory. This framework overcomes the shortcomings of traditional methods that ignore user differences and have coarse estimates of schedulable capacity. By realizing the simplification of the two-layer model through KKT conditions, the framework balances solution efficiency and global optimality, ensuring the lowest operating cost and the smallest peak-to-valley difference of the microgrid, significantly reducing grid pressure and improving the microgrid's ability to accommodate large-scale EVs.

[0013] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this specification. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments or the prior art 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 based on these drawings without creative effort.

[0015] Figure 1 This is a schematic diagram illustrating an application scenario of the coordinated charging scheduling method for electric vehicles considering differentiated charging needs under an islanded microgrid, as provided in this application embodiment. Figure 2 This is a flowchart illustrating the electric vehicle coordinated charging scheduling method that considers differentiated charging needs, provided in an embodiment of this application. Detailed Implementation

[0016] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0017] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0018] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0019] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0020] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0021] References to "one embodiment" or "some embodiments" in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0022] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] Reference Figure 1 The diagram illustrates an application scenario of the electric vehicle coordinated charging scheduling method for isolated microgrids, considering differentiated charging demands. This method is applied to scenarios in isolated microgrids characterized by large fluctuations in renewable energy output, significant peak-to-valley load differences, and differentiated electric vehicle charging demands. It proposes a master-slave game optimization strategy based on "three-state" schedulable capacity assessment and user charging urgency indicators for autonomous selection of fast / slow charging modes. Due to the dispersed location, small capacity, and high randomness of electric vehicles, directly incorporating them into the isolated microgrid control system is difficult. This necessitates the aggregation and bridging role of the communication and dispatch center. By aggregating dispersed clusters of various types of electric vehicles, the communication and dispatch center forms clusters with a certain scale and adjustment capabilities, making them easier for the isolated microgrid energy management system to schedule and control. Simultaneously, the dispatch center can formulate reasonable charging / discharging plans and power control schemes based on the real-time operating status of the isolated microgrid and charging pile resources to optimize the charging and discharging behavior of electric vehicles, achieving the goals of peak shaving and valley filling, reducing operating costs, and improving renewable energy absorption rates.

[0024] Figure 2 This is a schematic flowchart of an embodiment of the electric vehicle coordinated charging scheduling method provided in this application. (Refer to...) Figure 2 The scheduling method is described in detail below: Step 201: Obtain basic data for electric vehicles.

[0025] Basic data includes, but is not limited to: battery capacity, rated charging / discharging power, charging / discharging efficiency, real-time state of charge (SOC), on-grid / expected off-grid time, driving mileage, charging pile number, and historical charging preferences. The data is collected from multiple sources, including on-board terminals, smart charging piles, and transportation platforms, providing raw input for subsequent assessment of dispatchable potential and personalized scheduling.

[0026] Step 202: Construct a "three-state" schedulable capacity assessment model.

[0027] Electric vehicles are classified into three states based on their real-time State of Charge (SOC): "chargeable only", "dischargeable only", and "both charging and discharging are possible". A single vehicle minute-level schedulable capacity calculation framework is established by combining access status, travel plan, number of charging piles and power constraints. Furthermore, Monte Carlo simulation is used to aggregate massive numbers of EVs to generate a regional total schedulable capacity curve, so as to achieve accurate quantification of schedulable potential under multi-dimensional constraints of "vehicle-charging pile-road-network".

[0028] Step 203: Calculate the charging urgency index (CUI) and classify the fast charging / slow charging modes.

[0029] Based on remaining network time, current SOC, minimum travel power demand, and slow charging capability, a CUI index is proposed. When CUIi < 0, it is determined to be an emergency charging situation, and fast charging mode is forced. When CUIi ≥ 0, it is classified as slow charging mode, providing user-side decision variables for differentiated scheduling and two-layer game.

[0030] Step 204: Establish a two-layer optimization model for the "microgrid-EV" master-slave game.

[0031] The upper-layer microgrid, acting as the leader, aims to minimize system operating costs by deciding on generator output, energy storage charging and discharging, and charging pile resource allocation. The lower-layer EV cluster, acting as the follower, aims to minimize the peak-valley load difference of the microgrid by responding to price / power signals from the upper layer and optimizing fast / slow charging times and power. The upper and lower layers form a master-slave game structure through bidirectional "resource-load" coupling, balancing the microgrid's economics with user charging needs.

[0032] Step 205: Perform a single-layer transformation based on KKT conditions.

[0033] The optimality condition (KKT) of the lower-level convex optimization problem is transformed into upper-level constraints. The Big-M linearization complementary relaxation term is adopted, and the strong duality theorem is introduced to eliminate the nonlinearity of the objective. Finally, the two-level model is equivalently transformed into a single-level mixed integer linear programming (MILP) problem, which significantly reduces the solution complexity.

[0034] Step 206: Solve for the globally optimal scheduling scheme.

[0035] The high-performance MILP solver is called to solve the transformed single-layer model. After integer correction and KKT backtracking verification, the output is a globally optimal scheduling scheme that includes the fast / slow charging time, charging power, energy storage and generator output of each EV, achieving Pareto optimality with minimum microgrid overall cost and minimum peak-valley difference.

[0036] Step 207: Issue the charging plan and implement closed-loop control.

[0037] The optimal scheduling result is encapsulated into minute-level charging instructions, which are then sent to the vehicle through the local controller of the smart charging pile. Simultaneously, a rolling time-domain feedback mechanism is established to collect real-time data on the EV's actual SOC, changes in off-grid time, and network load deviations. This triggers rolling recalculation of the model and secondary correction of the instructions, forming a closed-loop control of "planning-execution-measurement-correction" to ensure the robustness and feasibility of the scheduling scheme under complex operating conditions.

[0038] This embodiment integrates "three-state" schedulable capacity assessment, CUI-driven fast / slow charging partitioning, master-slave game modeling, and KKT single-layer conversion technology within an isolated microgrid. This achieves global optimization of orderly charging for electric vehicles that takes into account users' differentiated charging needs, effectively shaving peaks and filling valleys, reducing microgrid operating costs, and improving the safety, economy, and low-carbon operation of microgrids in scenarios with a high proportion of electric vehicles connected.

[0039] In one embodiment, raw data from electric vehicles is presented as raw data for assessing schedulable potential.

[0040] Step 2021: Analyze the mapping relationship between SOC and schedulable capacity.

[0041] First, we analyze the relationship between SOC and charging / discharging power, and the relationship between SOC and the driving range of electric vehicles.

[0042] Regarding the relationship between SOC and charge / discharge power: If the charging power of the electric vehicle is constant at time t: in SOC i ( t +Δ t () is an electric vehicle i exist t +Δ t The charging status at any given time; SOC i ( t () is an electric vehicle i The charging state at time t; P c,i It is an electric car i The charging power; C i It is an electric car i Battery capacity; η c,i It is an electric car i Charging efficiency; Δ t This is the time interval, set to 1 minute.

[0043] If the discharge power of the electric vehicle is constant at time t: in, P f,i It is an electric car i The discharge power; η f,i It is an electric car i The discharge efficiency.

[0044] Regarding the relationship between SOC and electric vehicle driving range: From the perspective of the charging and discharging potential of electric vehicles connected to the power grid, the State of Charge (SOC) and battery capacity of the vehicle battery play a decisive role, and the driving range of the electric vehicle is a key factor of the SOC. In this embodiment, assuming that the battery charge of the electric vehicle decreases at a constant rate during driving, the driving range relationship between the current SOC of the electric vehicle and the SOC at the time of the last charge can be calculated using the following formula: in, The electric vehicle i's SOC after its last charge; SOC i ( t start,j () is the SOC of electric vehicle i at the time interval j after charging is completed; SOC i ( t start,j+1 ) is the SOC of electric vehicle i at interval j+1 after the power supply is fully charged; d is the driving distance of electric vehicle i between interval j and interval j+1; D is the driving distance of electric vehicle i when the battery is fully charged.

[0045] Step 2022: Analyze the three scenarios in which the SOC determines schedulable capabilities.

[0046] The real-time schedulable capacity of electric vehicles (EVs) within a given area is constantly changing. Therefore, the real-time State of Charge (SOC) of an EV is a crucial factor that cannot be ignored when calculating its real-time charging and discharging capacity. Despite travel plan constraints, the relationship between the EV's SOC and its potential should be considered first. Based on the EV's current SOC, this relationship can be categorized into three scenarios, as shown in Table 1, which represents three scenarios of schedulable capacity. Table 1. Three scenarios for schedulable capabilities Step 2023: Scheduling Capacity Assessment Algorithm Based on the preceding analysis, we can derive the behavioral and charging states of electric vehicles (EVs) from their travel time and distances, thus obtaining the scheduling potential of each EV. The scheduling potential is characterized by the maximum rechargeable amount of each EV, calculated as follows: in, C i,charge ( t ) is the maximum rechargeable amount of vehicle i at time t, which is affected by the current SOC and battery capacity C. i Charging efficiency ηc,i and SOC upper limit SOC i,max constraint.

[0047] Monte Carlo simulation of large-scale EV group behavior is used. Inputs include the total number of EVs, the proportion of EVs of different types, the battery capacity of each EV, charging / discharging power, charging start time, and constraints on the total number of charging stations. For each EV, its schedulable type is determined based on its State of Charge (SOC) and behavioral state. Finally, the charging / discharging potential of all EVs is summarized, yielding the expression for the total schedulable capacity curve for the region: In the formula, It represents the maximum chargeable amount of all vehicles in the region at time t. It is the total set of vehicles in the region at time t.

[0048] Because the number of electric vehicle charging facilities is limited, the number of electric vehicles charging cannot exceed the number of charging stations; therefore, the constraint is: Where N is the total number of electric vehicle charging stations in the area. N f This represents the number of electric vehicles that currently meet the conditions for discharge scheduling.

[0049] This embodiment constructs a scientific and precise model for assessing the schedulable capacity of electric vehicles by deeply analyzing the mapping relationship between State of Charge (SOC) and schedulable capacity, ensuring accurate quantification of the charging and discharging potential of electric vehicles under different battery states. First, starting from the relationship between SOC and charging / discharging power, the direct impact of battery state of charge on charging and discharging power is clarified: under constant charging power, SOC increases linearly with time, and charging efficiency and battery capacity jointly determine the charging rate; while during discharging, SOC gradually decreases with power output, and discharging efficiency also affects the rate of energy release. By introducing a discretized calculation method with a time interval Δt = 1 min, high-precision tracking of dynamic changes in SOC can be achieved, providing real-time data support for subsequent scheduling models.

[0050] Secondly, this embodiment analyzes the relationship between State of Charge (SOC) and electric vehicle range, assuming that battery power is consumed at a constant rate during driving, and thus establishes a linear mapping model between SOC changes and driving distance. This model can predict the vehicle's range and power consumption trend over a future period based on the vehicle's historical charging records and current SOC, providing a basis for determining the vehicle's dispatchability potential. Especially before the vehicle connects to the grid, the system can estimate the urgency and dispatchability window of its charging needs based on its mileage and power consumption, providing data support for the formulation of differentiated dispatch strategies.

[0051] Furthermore, this embodiment refines the relationship between SOC and dispatchable capability into three typical scenarios: dispatchable charging only, dispatchable discharging only, and dispatchable charging and discharging. Each scenario corresponds to a different SOC range and operating state constraints, comprehensively covering the dispatch potential of electric vehicles under different power levels. For example, in the low SOC range, the vehicle only has charging capability and cannot discharge; in the medium-high SOC range, the vehicle can flexibly switch between charging and discharging states according to grid demand; in the high SOC range, the vehicle mainly has discharging capability, which can be used for grid peak shaving or supporting microgrid operation. Through scenario-based classification, the system can accurately identify the dispatchable type of each vehicle, avoiding mis-dispatching and resource waste.

[0052] Regarding the schedulable capacity assessment algorithm, this embodiment employs the Monte Carlo simulation method to model and simulate the behavior of a large-scale electric vehicle (EV) population. By inputting key data such as the total number of EVs, vehicle model ratios, battery parameters, charging and discharging power, and the number of charging stations, the system can simulate the schedulable capacity distribution of vehicles under differentiated State of Charge (SOC) states and aggregate it to generate a regional-level schedulable capacity curve. This curve not only reflects the charging and discharging potential of the EV population during differentiated time periods but also considers the upper limit constraint of the number of charging stations on charging capacity, ensuring the feasibility and safety of the scheduling scheme under actual physical conditions.

[0053] In one embodiment, in step 203, a multi-stage modular charging model is constructed based on the electric vehicle charging model and the user's personalized charging needs, including: Step 2031: Define charging urgency indicators based on the electric vehicle charging model.

[0054] In the proposed electric vehicle charging scheduling model, all electric vehicles are scheduled based on their charging needs. This section defines a charging urgency index (CUI) to reflect the urgency of an EV's charging needs.

[0055] According to arrival time slot J i c and departure time slot J i dis The entire time slot for electric vehicles to connect to a microgrid can be calculated using the following formula: in, T i Rem This represents the number of remaining time periods during which the i-th electric vehicle can continue to be connected to the microgrid, and the charging behavior and charging scheduling strategy should be arranged within these remaining time periods.

[0056] In this embodiment, the CUI formula is defined as follows: in, P EV slow This refers to the slow charging power of electric vehicles. η EV This indicates the charging efficiency of electric vehicles. SOC i min This represents the lower bound of the State of Charge (SOC) when the i-th electric vehicle finishes charging. SOC i con This represents the State of Charge (SOC) when the i-th electric vehicle is connected to the microgrid. Cap EV bat Indicates the battery capacity of an electric vehicle. When... CUI i When CUI < 0, it indicates that the charging demand of the i-th electric vehicle is urgent. In contrast, when CUI i When the value is 0, it means that the charging demand of the i-th electric vehicle is not urgent.

[0057] Step 2032: Select the charging mode based on the charging urgency index.

[0058] Here, the charging mode selection is determined by... CUI The decision is as follows: in, P EV fast Indicates fast charging power. P EV slow This indicates the slow charging power. When... CUI i < At time 0, the i-th electric vehicle uses fast charging mode. CUI i ≥ At time 0, the i-th electric vehicle uses slow charging mode.

[0059] In step 203 of this embodiment, a quantifiable and executable personalized charging demand identification and response mechanism is constructed with "charging urgency" as the core. First, in step 2031, an innovative charging urgency index CUI is proposed. When CUIi < 0, it means that even if the vehicle is continuously charged slowly at maximum power, it cannot reach the minimum battery level threshold before the planned disconnection from the grid, and the system immediately determines it as "urgent"; conversely, if CUIi ≥ 0, it is considered "not urgent". This index uses a one-dimensional scalar to simultaneously characterize the three-dimensional information of "battery gap, time margin, and charging capacity", laying a quantitative foundation for subsequent hierarchical and graded scheduling.

[0060] In step 2032, CUI is further mapped to the charging mode selection: CUIi < 0 automatically triggers fast charging mode, and the vehicle is assigned a high-power charging pile to replenish the battery in the shortest time; CUIi ≥ 0 enters slow charging mode, and the vehicle uses only a small amount of power, which not only flexibly completes the replenishment over a longer period of time, but also achieves the optimal balance between microgrid peak-valley regulation and user experience. By directly writing "urgency" into the scheduling constraints, the system can complete mode switching in milliseconds without manual intervention, significantly reducing decision latency; at the same time, the difference in fast and slow charging power brings natural load stratification, allowing the microgrid to use the slow charging group for peak shifting and valley filling, while the fast charging group participates in second-level frequency regulation as an interruptible high-priority resource.

[0061] More importantly, the CUI metric is linked in real time with the battery model, road network information, and charging pile condition database: when a user temporarily modifies the off-grid time, or when the SOC estimation deviates, the i-th vehicle's... The SOC gap is dynamically refreshed, and the CUI is recalculated online accordingly. The system can immediately upgrade or downgrade the charging mode, avoiding resource waste caused by "too hasty" or "too slow" charging. CUI maximizes the utilization of charging facilities, extends battery life, and enables flexible and controllable grid load.

[0062] In one embodiment, in step 204, a two-layer optimization model of master-slave game in an islanded microgrid containing electric vehicles is established, including: Step 2041: The isolated microgrid, as the leader, serves as the upper layer of the master-slave game two-layer optimization model.

[0063] For example, the upper layer limits the decision-making space of the lower layer through resource constraints, and the lower layer affects the cost of the upper layer through charging load feedback. CUI i <0 (Vehicle i-charging emergency), even during peak electricity consumption, the lower level needs to select fast charging, leading to an increase in total load; however, during peak hours, the base load... P t load The current level is already high; with the addition of fast charging load, the upper level needs to start high-cost generators (such as gas turbines). ci (Large) or release stored energy (increase) c bat Costs), which will lead to the overall operating cost of the system. F grid rise.

[0064] The objective function is as follows: in, c i This represents the unit output cost of the i-th power generation unit. P i,t gen For the i-th generating unit to output power at time t, c bat Depreciation costs for charging and discharging energy storage units. P t bat,ch / dis For energy storage charging and discharging power, c loss The penalty cost per unit of power outage, P t load Total load power, P t loss The power loss is represented by T; the total time period is represented by T.

[0065] The constraints are as follows: Power balance constraints: Among them, P t EV This indicates the total amount of electricity charged for an electric vehicle.

[0066] Electric vehicle charging constraints: Upper and lower limits of power generation output constraints: In the formula, and These are the minimum and maximum outputs of the i-th unit, respectively.

[0067] Fundamental constraints of energy storage: In the formula, Let t be the amount of electricity stored in the energy storage system at time t. This represents the maximum energy storage capacity of the energy storage system.

[0068] Charging station resource constraints: In the formula, The state variable representing the fast charging of the i-th vehicle at time t; , These represent the collections of fast-charging vehicles and slow-charging vehicles within the region, respectively. The number of fast charging stations in the area. This refers to the number of slow-charging stations in the area.

[0069] Step 2042: Establish an electric vehicle scheduling layer that takes into account charging urgency indicators.

[0070] The optimization goals are as follows: In the optimization model, state variables x i,j It is a binary variable. If... x i,j =0, indicating that the i-th electric vehicle was not charging in the j-th time slot; if x i,j If 1 = 1, it means that the i-th electric vehicle is charging in the j-th time slot, as shown in the following formula: The total load of the microgrid includes baseload and electric vehicle (EV) charging load, which in turn includes fast-charging EV load and slow-charging EV load. Therefore, the total load in the j-th time slot of the cooperative scheduling model... It can be represented as: in P con j This represents the basic load during the j-th time slot. This represents the power provided to all electric vehicles in the j-th time slot. P EV,i It is determined by the formula in step 2: if the i-th EV has an emergency charging need. P EV,i equal P EV fast If the i-th EV does not have an emergency charging need, then P EV,i equal P EV slow .

[0071] The goal of charging scheduling is to reduce the peak-valley load difference in a microgrid, which can be expressed as: in P T-c max and P T-cmin These represent the maximum and minimum load requirements, respectively.

[0072] The constraints are as follows: Step 2043: Analyze the constraints of fast-charging electric vehicles based on the lower-level objective function.

[0073] The time slot for stopping charging of electric vehicles is determined by the vehicle's departure time and the time slot required to meet maximum SOC (State of Charge) demand. The restrictions on emergency-charging electric vehicles can be expressed as follows: The fast charging power cannot exceed the maximum dispatchable power limit for a single vehicle determined in step 1. in, J i end The time slots during which fast-charging electric vehicles stop charging. J i end It can be represented as: in, SOC i max This is the maximum SOC demand. This indicates the charging time slots for electric vehicles that need to meet maximum SOC requirements. This indicates an integer less than the result of the division operation, which avoids overcharging of the EV in the next integer.

[0074] Step 2044: Analyze the constraints of slow-charging electric vehicles based on the lower-level objective function.

[0075] Due to the diverse charging needs of electric vehicle owners, each electric vehicle has a minimum SOC requirement and a maximum SOC requirement to avoid overcharging. For electric vehicles charging slowly, when the vehicle is disconnected from the microgrid, its minimum SOC requirement must be met, as expressed below: in, This is the SOC when the i-th EV is disconnected from the microgrid, and it can be calculated in the following way: in, This includes the total number of slow-charging electric vehicles. The second limitation of slow-charging electric vehicles relates to scheduling time. Charging activity can only be scheduled while the electric vehicle is connected to the microgrid. Therefore, when a slow-charging electric vehicle is not connected to the microgrid, the charging state... It must equal 0. It's important to note that the scheduling plan cannot be executed when the electric vehicle is disconnected from the microgrid.

[0076] Step 2045: Analyze the constraints of the microgrid based on the lower-level objective function.

[0077] The objective function is to reduce the peak-to-valley difference of the total load, but it cannot guarantee that the coordinated peak value will be lower than the peak value of the uncoordinated charging method that meets the maximum SOC requirement of electric vehicles. To avoid new peak charging loads in the microgrid during the coordinated scheme, the constraints on the microgrid are as shown in the formula: in, The peak-to-valley difference of the total load; It is the maximum total load in the non-coordinated charging mode that meets the maximum SOC demand of electric vehicles within 96 time slots, that is, the peak total load of the microgrid in the globally optimal scheduling scheme is not greater than the maximum charging load in the non-scheduled scenario.

[0078] In step 204 of this embodiment, a two-layer optimization model of "microgrid-EV" master-slave game is constructed: the upper-layer islanded microgrid aims to minimize operating costs, and limits the decision space of the lower layer through constraints such as power balance, generator output, energy storage and charging pile resources, and bears the high-cost generator or energy storage actions brought about by the fast charging demand of the lower layer; the lower-layer EV scheduling aims to reduce the peak-valley difference of the microgrid, and divides vehicles into fast charging and slow charging categories according to the CUI index, and sets constraints such as maximum dispatchable power, upper and lower limits of SOC, and access time period respectively, to ensure that fast charging vehicles complete the charging in the shortest necessary time and slow charging vehicles flexibly fill the gap during the grid stay period. At the same time, a mandatory constraint of "the peak value after coordination shall not be higher than the uncoordinated peak value" is applied to the microgrid side to achieve a closed-loop win-win situation of peak shaving and valley filling and charging demand satisfaction.

[0079] In one embodiment, in step 205, the two-layer model is transformed and solved using KKT conditions: To simplify the solution process, this embodiment constructs the Lagrangian function of the lower-level model and introduces KKT conditions, the Big-M method, and duality theory to linearize the model, transforming the problem into a single-level mixed-integer linear programming problem. First, the Lagrangian function of the lower-level model is constructed and equivalently transformed based on KKT conditions, and the Big-M method is used to linearize the complementary relaxation conditions. Then, the nonlinear terms in the upper-level objective function are eliminated according to the strong duality theorem, ultimately transforming the model into a mixed-integer linear programming problem.

[0080] Step 206 solves the transformed model using a convex optimization solver to obtain a relaxed solution. Then, the solution is adjusted by integerization and backtracked to verify the optimality of the KKT conditions to obtain the globally optimal scheduling scheme. Specifically, the transformed single-layer MILP model is fed into a high-performance solver such as Gurobi / Cplex, and the globally optimal charging-generating plan that takes into account the microgrid cost and peak-valley difference is obtained through KKT backtracking verification.

[0081] Step 207 encapsulates the aforementioned globally optimal scheduling scheme into minute-level OCPP instructions and sends them to the charging piles. It also uses rolling MPC closed-loop correction to adjust SOC and off-grid deviations, achieving robust control through "planning-execution-measurement-correction" to ensure the continued feasibility of the peak-shaving and valley-filling scheme in a real islanded microgrid. This application proposes an "orderly charging scheduling method for electric vehicles considering differentiated charging needs in an islanded microgrid environment." First, based on the mapping relationship between real-time SOC and mileage of electric vehicles, vehicles are divided into three schedulable capacity models: "chargeable only," "dischargeable only," and "both charging and discharging are possible." Monte Carlo simulation is used to aggregate regional-level minute-level adjustable potential curves. Subsequently, using remaining on-grid time, minimum travel energy, and slow charging capability as variables, a charging urgency index CUI is constructed. Fast charging / slow charging modes are automatically allocated based on CUI < 0 or ≥ 0, quantifying differentiated user-side needs. Furthermore, a two-layer optimization framework of "microgrid-EV" master-slave game is established: the upper-layer microgrid aims to minimize operating costs, deciding on generator, energy storage, and charging pile resources; the lower-layer EV cluster aims to minimize peak-to-valley differences, responding to upper-layer price / power signals and optimizing fast / slow charging times and power. By using KKT conditions, the lower-layer convex optimization problem is transformed into upper-layer constraints, completing the simplification of the two-layer model into a mixed-integer linear programming problem, and a high-performance solver is invoked to obtain the globally optimal scheduling scheme. This embodiment effectively smooths peaks and fills valleys while balancing the economical operation of the microgrid with users' personalized charging needs, improving the safety, economy, and low-carbon operation level of isolated microgrids in scenarios with a high proportion of electric vehicles.

[0082] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0083] It should be noted that although several units / modules or sub-units / modules of the microgrid electric vehicle orderly charging scheduling method technology considering differentiated charging needs have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.

[0084] Furthermore, although the operations of the method of this application are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0085] This application also provides a computer-readable storage medium storing computer-executable instructions. When a processor executes the computer-executable instructions, it implements the microgrid electric vehicle orderly charging scheduling method considering differentiated charging needs provided in the above embodiments of this application.

[0086] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0087] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for orderly charging scheduling of microgrid electric vehicles considering differentiated charging demands, characterized in that, include: Acquire basic data on various types of electric vehicles; Based on the aforementioned basic data, a schedulable capacity assessment model is constructed to calculate the schedulable capacity under differentiated charging modes; Based on the schedulable capacity and charging urgency indicators, a master-slave game two-layer optimization model is constructed with the microgrid as the leader and electric vehicle charging as the follower. Based on the KKT conditions, the master-slave game two-layer optimization model is transformed into a single-layer mixed integer linear programming problem, and the globally optimal scheduling scheme is solved.

2. The method as described in claim 1, characterized in that, Constructing the schedulable capacity assessment model includes: Based on the mapping relationship between SOC, charging and discharging power, and driving range, three scheduling scenarios are defined: electric vehicles can only be charged, electric vehicles can only be discharged, and electric vehicles can be charged and discharged.

3. The method as described in claim 1, characterized in that, The process of solving for the schedulable capacity includes: The behavioral and charging states of electric vehicles are obtained based on their driving time and corresponding distance, and the scheduling potential of each electric vehicle is calculated. This scheduling potential is characterized by the maximum rechargeable amount of each electric vehicle, and the formula is as follows: in, C i,charge ( t ) is the maximum chargeable amount of vehicle i at time t, affected by the current Battery capacity C i Charging efficiency η c,i and SOC upper limit SOC i,max constraint; The Monte Carlo simulation is used to model the group behavior of electric vehicles. The inputs include at least the total number of electric vehicles, the proportion of different types of electric vehicles, the battery capacity of each electric vehicle, the charging and discharging power, the charging start time, and the constraints of the total number of charging stations. Based on the SOC and behavioral state of each electric vehicle, the corresponding schedulable type is determined. The total dispatchable capacity curve of the region is obtained by summing up the dispatch potential of all electric vehicles in the region.

4. The method as described in claim 3, characterized in that, The formula for the charging urgency index CUI is: in, P EV slow This refers to the slow charging power of electric vehicles. η EV This indicates the charging efficiency of electric vehicles. SOC i min This represents the lower bound of the State of Charge (SOC) when the i-th electric vehicle finishes charging. SOC i con This represents the State of Charge (SOC) when the i-th electric vehicle is connected to the microgrid. Cap EV bat Indicates the battery capacity of an electric vehicle; Δ t It is a time interval. T i Rem This represents the remaining time period during which the i-th electric vehicle can continue to be connected to the microgrid.

5. The method as described in claim 4, characterized in that, The master-slave game model includes: the upper microgrid layer aims to minimize the overall system operating cost, and sets constraints on power balance, upper and lower limits of power generation output, energy storage operation, and charging pile resources; The lower-level electric vehicle scheduling layer aims to minimize the peak-valley load difference of the microgrid. It divides the charging urgency index into fast charging mode or slow charging mode, and satisfies the upper and lower limits of charging amount and the access time matching constraint.

6. The method as described in claim 5, characterized in that, When the charging urgency index CUI < 0, the vehicle adopts the fast charging mode; when the charging urgency index CUI ≥ 0, the vehicle adopts the slow charging mode.

7. The method as described in claim 1, characterized in that, The model transformation based on KKT conditions includes: constructing the Lagrangian function of the lower-level model, introducing KKT conditions, the Big-M method, and the strong duality theorem, and transforming the two-level problem into a single-level mixed integer linear programming problem.

8. The method as described in claim 7, characterized in that, The model solution based on KKT conditions includes: The relaxed solution is obtained by using a convex optimization solver. After integer adjustment and backtracking the KKT conditions to verify its optimality, the globally optimal scheduling scheme is obtained.

9. The method as described in claim 1, characterized in that, The globally optimal scheduling scheme satisfies: The peak total load of the microgrid under the globally optimal scheduling scheme is no greater than the maximum charging load in non-scheduled scenarios; When a vehicle is off-grid, its SOC (State of Charge) should not be lower than the minimum SOC requirement set by the user.

10. The method as described in claim 1, characterized in that, The schedulable capacity assessment model also includes a limit on the number of charging piles, with the following constraints: Where N is the total number of charging stations in the area. N f This represents the number of electric vehicles that currently meet the conditions for discharge scheduling.

Images