Electric vehicle fast and slow charging scheduling optimization method based on dynamic urgency assessment

Abstract

The invention discloses an electric vehicle fast and slow charging scheduling optimization method based on dynamic urgency evaluation, and aims to solve the problems of static urgency judgment, no quantization of battery loss, poor peak-valley coordination and low large-scale scheduling efficiency in existing scheduling. According to the method, charging data of the electric vehicle are collected, a user charging time normal distribution model is constructed, and time slot discretization processing is implemented; an emergency comprehensive index CUI is calculated, a dynamic threshold value is set according to peak / flat / valley time period characteristics of the power grid, and accurate judgment of a charging mode is achieved; and load superposition optimization is carried out by taking minimization of the peak-valley difference of the micro-grid as a target, meanwhile, a fast and slow charging mode switching constraint, a battery health state constraint and a micro-grid safe operation boundary are integrated, and an optimal charging period distribution scheme is obtained by utilizing a CPLEX solver. The peak-valley difference of a power grid is effectively reduced, the fast charging loss of a battery is reduced, the charging satisfaction degree of a user is improved, minute-level response large-scale vehicle group scheduling is realized, and the method is particularly suitable for public charging scenes such as office parks and commercial complexes.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle charging scheduling and optimization technology, and in particular to an optimization method for fast and slow charging scheduling of electric vehicles based on dynamic urgency assessment. Background Technology

[0002] Against the backdrop of the accelerated construction of new power systems, the large-scale integration of electric vehicles (EVs) has become an irreversible trend. While bringing significant environmental benefits, this trend also poses a severe challenge to the peak-valley regulation capabilities of microgrids. The spatiotemporal aggregation effect of EV charging load is significant, especially in public charging scenarios. The disorderly charging behavior of a large number of EVs can easily lead to an expansion of the peak-valley difference in grid load, affecting the safe and stable operation of the grid.

[0003] Current research on electric vehicle charging scheduling mainly focuses on the dynamic allocation of charging modes and real-time load matching. Some studies attempt to improve the smoothness of the grid load curve by establishing fast and slow charging priority scheduling mechanisms combined with Monte Carlo stochastic simulation methods. However, these methods still have significant shortcomings in terms of real-time response.

[0004] First, existing technologies have significant shortcomings in quantifying battery loss constraints during fast charging. For example, while some technologies, such as CN111626527A, which considers the deep learning scheduling method for smart grids in scheduling fast / slow charging and discharging modes of electric vehicles, propose scheduling strategies that take into account the charging and discharging power of electric vehicles, they fail to fully quantify battery loss during fast charging. This exacerbates the risk of equipment operating beyond its capacity, severely impacting battery lifespan and equipment safety in the long run. Furthermore, the lack of dynamic monitoring and quantitative constraint mechanisms for battery health further limits the ability of existing technologies to protect batteries and extend equipment lifespan.

[0005] Secondly, the modeling accuracy of the coupling between slow charging power allocation and real-time user access status is insufficient. Existing technologies, such as CN117465279A, disclose an intelligent control system and method for peak charging power of electric vehicles. Although adaptive allocation of charging power is achieved through big data analysis, in slow charging scenarios, the diversity and uncertainty of user charging behavior make it difficult to achieve optimal spatiotemporal allocation of slow charging power. This results in low utilization of charging capacity during off-peak hours, failing to fully realize the potential of slow charging in balancing grid load.

[0006] More importantly, traditional static optimization models lag in tracking the grid's base load, making it difficult to dynamically match charging load with the grid's carrying capacity. During periods of surging emergency charging demand, the lack of a tiered cutoff mechanism based on real-time load thresholds can easily lead to localized overloads, further exacerbating grid instability. Simultaneously, the power surges caused by charging state transitions are not incorporated into the existing technological constraints, potentially leading to secondary problems such as harmonic pollution during actual dispatching, thus affecting the grid's power quality.

[0007] Furthermore, current technical bottlenecks are mainly manifested in the following aspects: insufficient coordinated optimization between urgency assessment indicators and real-time grid operation data; the accuracy of demand grading restricts the effectiveness of scheduling strategies; the low efficiency of solving large-scale electric vehicle fleet mixed-integer programming models cannot meet the needs of minute-level scheduling instruction generation; and the lack of coupled modeling between overload protection mechanisms for fast-charging equipment and battery health status, which exacerbates equipment wear and tear over long-term operation. These problems collectively make it difficult for existing optimization methods to fully realize the potential of real-time scheduling, and there is an urgent need to construct a new control framework that integrates dynamic urgency assessment and multi-timescale load matching.

[0008] Regarding peak-valley coordination, existing technologies mostly employ simple peak-valley time-sharing strategies, concentrating charging loads during off-peak hours to reduce pressure on the grid during peak periods. However, this strategy does not fully consider the diversity and randomness of user charging demands, resulting in some urgent charging needs not being met in a timely manner and potentially causing grid overload during off-peak hours. Furthermore, with the rapid growth of electric vehicle ownership, charging scheduling for large-scale vehicle fleets has become a new challenge. Existing technologies often face high computational complexity and low solution efficiency when scheduling on a large scale, making it difficult to achieve real-time scheduling responses within minutes.

[0009] In summary, existing electric vehicle charging scheduling technologies have significant shortcomings in areas such as urgency assessment, battery loss constraints, peak-valley coordination, and large-scale scheduling efficiency, making it difficult to meet the actual needs of electric vehicle charging scheduling in microgrid environments. Therefore, developing an electric vehicle fast and slow charging scheduling optimization method based on dynamic urgency assessment has significant practical implications and application value. This method needs to be able to respond to grid load changes in real time, accurately quantify battery loss, optimize peak-valley coordination strategies, and achieve efficient scheduling of large-scale vehicle groups to address the challenges brought about by the large-scale integration of electric vehicles. Summary of the Invention

[0010] The technical problem this invention aims to solve is to provide an electric vehicle (EV) fast / slow charging scheduling optimization method based on dynamic urgency assessment, addressing the decline in peak-valley regulation capacity caused by the spatiotemporal aggregation effect of charging load during the large-scale integration of EVs into microgrids. Specifically, existing technologies suffer from the following limitations: ambiguous classification of emergency charging demand leads to scheduling inaccuracies; the lack of battery loss constraints during fast / slow charging mode switching poses equipment overload risks; the difficulty in coordinating grid peak-valley regulation with differentiated user needs; and the inefficiency of coupled modeling of random charging behavior and deterministic constraints in large-scale vehicle groups. This invention aims to overcome these technical bottlenecks and achieve dynamic matching between charging load and grid carrying capacity.

[0011] To achieve the above-mentioned technical objectives, the present invention adopts the following technical solution: This invention constructs a complete fast-slow-charge scheduling optimization system through a technical approach of "mathematical modeling - dynamic evaluation - load optimization - multi-constraint solution". The specific technical solution is as follows: (a) Collect electric vehicle charging data and perform mathematical modeling of the EV system. Collect charging data for all electric vehicles, including current battery level, charging time limit, minimum state of charge requirement, and maximum state of charge requirement, and initialize counters to prepare for processing each vehicle individually. To address the user behavior characteristics in public charging scenarios (such as office areas), a system model incorporating "charging time windows" and "time discretization" is constructed to provide a quantitative basis for subsequent scheduling. User charging time model: Considering the arrival time of electric vehicles (connection to the microgrid) Time of departure (disconnection from microgrid) Both exhibit randomness and follow a normal distribution, as expressed by: (1); (2); in, , These are the expected arrival / departure times, respectively. , These represent the standard deviations of arrival and departure times, aligning with the user behavior pattern of "arriving at the office early and leaving late" in public settings.

[0012] Time Discretization Model: To improve scheduling real-time performance, a day is divided into 96 time slots (each time slot is 15 minutes). The time slot number represents the time node when a vehicle connects to or disconnects from the microgrid. The expression is as follows: (3); (4); in, The total number of electric vehicles. For vehicle indexing, , These are the access / disconnection time slot numbers, The time slot length, (Round up) (Round down) to ensure precise matching of time nodes and time slots.

[0013] (II) Dynamic urgency assessment and pattern determination of charging demand A Critical Urgency Index (CUI) is introduced to quantify the urgency of charging needs, and the judgment threshold is dynamically adjusted in combination with the characteristics of the power grid during different time periods to achieve differentiated allocation of fast and slow charging modes: Remaining charging time slots are calculated: Based on the access / disconnection time slots, the number of remaining charging time slots for the vehicle is determined, providing a time basis for urgency assessment. (5); Emergency Indicator (CUI) Construction: By coupling slow charging power, charging efficiency, SOC (State of Charge) requirement, and battery capacity, it determines whether slow charging meets the demand. The expression is: (6); in, , These are the fast charging and slow charging powers, respectively. For charging efficiency; , These are the minimum SOC at the end of charging and the initial SOC upon connection, respectively. The value represents the battery capacity; the smaller the CUI value, the more difficult it is for slow charging to meet the demand, and the higher the urgency.

[0014] Time-based dynamic threshold adjustment: Based on the differences in grid carrying capacity during peak, stable, and off-peak periods, different CUI (Customer User Experience) judgment thresholds are set to achieve coordination of "time period-demand-mode". Peak hours (high grid load): Determined to be an urgent need, fast charging was activated; Stable periods (moderate grid load): Determined to be an urgent need, fast charging was activated; Off-peak hours (low grid load): Determined to be an urgent need, fast charging was activated; If a vehicle's access time slot spans multiple time periods, the time period with the largest proportion of its access time is taken as the dominant time period, and the corresponding threshold is used to avoid judgment errors caused by time period switching.

[0015] Charging mode and power determination: via binary variables Identify emergency situations and then allocate charging power accordingly: Emergency status determination: (7); For time period thresholds, ; Charging power allocation: (9); Charging status management: The vehicle's charging status is identified by formula (10) at the time of charging. The charging status of a time slot provides a basis for subsequent scheduling: (10).

[0016] (III) Load superposition and optimization target construction With the core objective of "minimizing the peak-valley load difference of the microgrid," the goal is to achieve coordinated matching between charging load and base load. Total load calculation: The total load of the microgrid includes the base load and the electric vehicle charging load (fast charging + slow charging), expressed as follows: (11); in, For the first Total load of each time slot; For the first Basic load for each time slot; Indicates the first EVs in the The charging status over a period of time; Indicates the first Charging power of each vehicle; For all vehicles in the first Total charging power of the time slot.

[0017] Optimization objective setting: To smooth the power grid load curve by minimizing the difference between the maximum and minimum total load, the expression is: (12); in, , These are the maximum and minimum values ​​of the total load within the 96 time slots, respectively.

[0018] (iv) Construction of multiple constraints and solution of optimal scheduling To ensure charging safety, equipment stability, and grid reliability, a multi-dimensional constraint model for fast and slow charging vehicles and microgrids is constructed, and the optimization model is solved using the CPLEX tool: Fast-charging vehicle constraints: Balancing emergency charging needs with battery safety, three constraints are set: Charging status constraints: Emergency vehicles ( ) Continuous charging during the time slot from connection to termination of charging, i.e.: (13); The charging stop time slot constraint is determined by the off-grid time, maximum SOC demand, and maximum battery loss, and is expressed as follows: (14); in, To meet the maximum SOC requirements, The maximum loss coefficient for a single fast charge of the battery, avoiding overcharging and excessive loss; Cumulative energy constraint: The total energy of fast charging shall not exceed the upper limit of the battery's allowable energy loss, expressed as: (15).

[0019] Slow-charging vehicle constraints: Ensuring basic user charging needs and rational scheduling. SOC constraint: When the microgrid is disconnected, the SOC must satisfy "minimum demand ≤ SOC ≤ maximum demand", that is: (16); in (State of Charge (SOC) upon disconnection) is calculated using the following formula: (17); Time constraint: When the vehicle is not connected to the microgrid ( Charging status This helps avoid ineffective scheduling.

[0020] Microgrid constraints: To prevent coordinated dispatch from triggering new load peaks, the constraints are as follows: (18); in, The maximum total load for "uncoordinated charging that meets the maximum SOC demand of electric vehicles" within 96 time slots is set to ensure that the peak load of the power grid does not increase after dispatching.

[0021] Optimization Solution: The CPLEX tool is used to solve the above multi-constraint mixed integer optimization model. Its parallel computing capabilities are utilized to generate dispatch instructions for large-scale electric vehicle groups (such as hundreds of vehicles) in minutes, providing data support for real-time power grid dispatch.

[0022] The electric vehicle fast and slow charging scheduling optimization method proposed in this invention, based on dynamic urgency assessment, has the following beneficial effects: 1. This invention effectively solves the problem of reduced peak-valley regulation capacity caused by the spatiotemporal aggregation of charging loads when electric vehicles are connected to microgrids on a large scale, and provides a practical technical solution for the optimized scheduling of microgrids.

[0023] 2. This invention overcomes the scheduling inaccuracy problem caused by the ambiguity of emergency charging demand classification in the prior art, and realizes dynamic matching between charging load and grid carrying capacity.

[0024] 3. This invention overcomes the risk of device overload caused by the lack of battery loss constraints when switching between fast and slow charging modes. It quantifies the fast charging battery loss constraint coefficient and cumulative energy limit to avoid device overcapacity and battery overcharging.

[0025] 4. This invention solves the contradiction between the difficulty in coordinating peak-valley regulation of the power grid and the differentiated needs of users. It dynamically matches the power grid carrying capacity with user needs based on time periods and improves user charging satisfaction.

[0026] 5. This invention overcomes the problem of low efficiency in modeling the coupled random charging behavior and deterministic constraints of large-scale vehicle groups by using Monte Carlo simulation combined with mixed integer linear programming to improve the modeling efficiency of random scenarios.

[0027] 6. This invention proposes a dynamic urgency assessment model, which establishes a quantitative system of urgency indicators based on battery remaining capacity, user charging time window constraints, and grid operating characteristics, and realizes the classification of charging modes.

[0028] 7. This invention constructs a multi-time-period mixed integer optimization model with minimizing the peak-valley difference of the system as the core objective, integrating fast-charging battery loss quantification constraints, slow-charging power spatiotemporal allocation strategies, and microgrid load safety boundaries.

[0029] 8. This invention utilizes Monte Carlo simulation to optimize the time period distribution of conventional slow charging, and combines mixed integer linear programming to solve the optimal scheduling scheme for multiple time periods, thereby realizing dynamic collaborative control in public charging scenarios.

[0030] 9. This invention constructs a fast / slow charging priority truncation mechanism and a minute-level time window discretization model, integrates the parallel computing capabilities of the CPLEX solver, and realizes real-time generation and dynamic tracking of minute-level scheduling instructions.

[0031] 10. This invention develops a parallel computing framework based on the CPLEX solver, which meets the minute-level scheduling requirements and improves the scheduling response speed by more than 30% compared with the traditional model, thus significantly improving scheduling efficiency.

[0032] 11. This invention limits emergency fast charging loads within the grid carrying capacity threshold through dynamic urgency assessment and multi-time period mixed integer optimization, thereby reducing the system peak-valley difference, mitigating the risk of local overload, and improving grid stability.

[0033] 12. This invention improves the charging capacity utilization rate to 88% during off-peak hours through a power adaptive allocation mechanism, reduces resource waste, and increases the user demand satisfaction rate to over 95%.

[0034] 13. User feedback indicates that the present invention provides dual protection for charging efficiency and device safety, and improves the battery health maintenance rate to over 92%, thus solving the coupling problem of device overload and battery health decline during fast charging.

[0035] 14. This invention combines fast and slow charging electric vehicles and microgrid constraints, and uses the CPLEX tool to calculate the optimal charging time period, providing data support for grid dispatching and enabling real-time generation of large-scale vehicle group dispatching instructions.

[0036] 15. This invention effectively balances the relationship between "grid safety, equipment lifespan, and user needs," and solves the harmonic pollution problem caused by the lag in tracking the grid base load and power surges caused by charging state switching in traditional static optimization models.

[0037] 16. This invention reduces the risk of system peak-valley difference and equipment overload. By establishing a dynamic evaluation index for charging urgency and a multi-period mixed integer optimization model, the emergency fast charging load is limited to the grid carrying capacity threshold. At the same time, Monte Carlo simulation is used to optimize the time distribution of conventional slow charging, thereby achieving spatiotemporal decoupling between charging load and base load.

[0038] 17. This invention improves user satisfaction and battery health by introducing quantitative constraints on battery loss during fast charging and an adaptive allocation mechanism for slow charging power. It ensures that emergency vehicles can be fully charged quickly while reducing the impact of high-rate charging on battery life, thus balancing charging efficiency and equipment safety.

[0039] 18. This invention improves the operational stability and dispatch response speed of microgrids. By constructing a fast and slow charging priority cutoff mechanism and a minute-level time window discretization model, and integrating the parallel computing capabilities of the CPLEX solver, it enables real-time generation and dynamic tracking of large-scale vehicle group dispatch instructions. Attached Figure Description

[0040] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 This is a block diagram of the electric vehicle charging scheduling model of the present invention; Figure 2 This is a flowchart illustrating the calculation process of the electric vehicle coordinated charging method that takes into account urgent needs according to the present invention. Figure 3 This is the total load calculation process for the uncoordinated charging method of the present invention; Figure 4 This is a total load diagram for different scheduling strategies of the present invention; Figure 5 This is a slow charging load diagram for different scheduling strategies of the present invention; Figure 6 This is a fast-charging load diagram for different scheduling strategies of the present invention. Detailed Implementation

[0041] The technical solutions of the present invention will be further described below with reference to the embodiments and accompanying drawings: Example 1 like Figures 1 to 6 As shown in the figure, this embodiment provides a method for optimizing fast and slow charging scheduling of electric vehicles based on dynamic urgency assessment. The specific implementation steps and processes are as follows, in conjunction with the accompanying drawings: Figure 1 The diagram shown is a framework diagram of an embodiment of the electric vehicle charging scheduling method considering urgent needs provided by the present invention. The framework diagram is as follows: In this model framework, when an EV owner connects their electric vehicle to a charging station, they upload their charging demand information to the dispatch center. This information includes the EV's arrival and departure times, the State of Charge (SOC) at the time of connection, and the SOC required to meet the EV's charging needs. Based on the collected information, the dispatch center will formulate a charging and discharging plan for each EV and send it to the corresponding EV user. In the non-coordinated charging method, once an EV is connected to a charging station, it begins charging continuously until the SOC required to meet the charging needs is reached or the predetermined departure time is reached. In the coordinated charging method, an indicator reflecting the urgency of charging is introduced to distinguish whether an EV's charging demand is urgent, categorizing it as urgent or non-urgent. EVs with urgent charging needs are charged using fast charging, while other EVs are charged using slow charging.

[0042] Figure 2 This is a calculation flowchart of the electric vehicle coordinated charging method of the present invention that takes into account urgent needs. The flowchart specifically includes the following steps: Step 1: Perform mathematical modeling on each part of the electric vehicle system, and construct a trolley system that includes the charging time window of users in public scenarios and time discretization.

[0043] Specifically, the mathematical models for each part are established through the following steps: A user charging time model is established, considering the behavior of electric vehicle owners in public areas. Under public charging conditions, electric vehicle owners begin charging when they arrive at their workplace in the morning and finish charging after get off work. Electric vehicle arrival time... and departure time It also follows a normal distribution, which can be represented as: (1); (2); In the formula, and These are the expected arrival and departure times for the public charging mode, respectively. and These are the standard deviations of arrival and departure times for public charging modes.

[0044] During the scheduling process, scheduling plans are typically executed in time slots to improve efficiency. The scheduling time is divided into several time slots. In the proposed electric vehicle charging scheduling model, a day is discretized into 96 time slots, each slot being 15 minutes long. The arrival and departure times of each electric vehicle can then be expressed as: (3); (4); In the formula, Indicates the number of EVs. Indicates the first The index of electric vehicles, with a value range of 100. , Indicates the first The time slot number for each electric vehicle to connect to the microgrid. Indicates the first The time slot number when an electric vehicle disconnects from the microgrid. Indicates the first The arrival time of the electric vehicle (i.e., the time of connection to the microgrid). Indicates the first The time when an electric vehicle leaves (i.e., the time when the microgrid is disconnected). This indicates the length of a single time slot, which is 15 minutes in this case. Indicates the first Arrival time of electric vehicles Divide by the length of the time slot Round the result up. Indicates the first departure time of electric vehicles Divide by the length of the time slot The result is rounded down.

[0045] Step 2: Calculate the urgency index of each vehicle and dynamically adjust the threshold according to the peak, off-peak and stable periods of the power grid to determine the urgency of the vehicle's charging needs: for vehicles in urgent need, fast charging is directly activated until the vehicle is fully charged; for vehicles in non-urgent need, the demand is temporarily stored until all vehicles have been traversed.

[0046] Specifically, step 2 includes: According to arrival time slot and departure time slot The entire time slot for electric vehicles to connect to the microgrid can be calculated as follows: (5); In the formula, For the first The remaining number of time periods during which electric vehicles can continue to connect to the microgrid, and the charging behavior and charging scheduling strategies during these remaining time periods.

[0047] In the EV coordinated optimization scheduling model proposed in this invention, scheduling is performed based on the charging needs of EVs. An urgency index is introduced in this process. This metric reflects the urgency of EV charging needs. It indicates whether using a slow charging method while the EV is connected to a charging station can meet the target charging requirements. Furthermore, by identifying peak, off-peak, and stable periods in the power grid, the threshold for determining the vehicle urgency index (CUI) is dynamically adjusted: during peak periods... If the need is deemed urgent, fast charging will be activated; otherwise, slow charging will be used. During off-peak hours... Use fast charging only during peak hours, otherwise use slow charging; during stable periods... Fast charging is enabled when needed, and slow charging is enabled otherwise, thus achieving time-based differentiated charging optimization and control.

[0048] When the time slot for an electric vehicle to access the microgrid spans multiple grid time periods, the time period with the highest proportion of the time period during which the vehicle accesses is taken as the dominant time period, and the corresponding urgency threshold is applied.

[0049] At the same time, use binary variables This indicates whether the charging needs of each EV are urgent. When, it indicates the first EVs with urgent charging needs will use fast charging; conversely, when... When, it indicates the first If the EV's charging needs are not urgent, a slow charging method can be used. The calculation formula is as follows: (6); (7); (8); (9); In the formula, This indicates the slow charging power of the EV. For the fast charging power of EVs, This indicates the charging efficiency of the EV. Indicates the first EV connected to charging station , For the first When the EV finishes charging The lower bound, Indicates the battery capacity of the EV. This is a dynamic threshold parameter for a given time period. It refers to the time period in which the EV is located. The charging power for each EV, For fast charging power, This is the slow charging power.

[0050] In this model, charging state and discharging state represent the charging and discharging behavior of the EV, respectively. Therefore, coordinated charging and discharging scheduling of the EV is achieved by determining its charging or discharging state at various time periods. The expression is shown below: (10); In the formula, Indicates the first EVs in the The charging status over a period of time.

[0051] Step 3: Superimpose the fast and slow charging loads of all vehicles with the basic power grid load, and schedule them with the core objective of minimizing the power grid load difference.

[0052] Total load of microgrid It includes baseload and electric vehicle charging load, with the electric vehicle charging load consisting of fast charging load and slow charging load. Therefore, the coordinated charging scheduling model in the first... The total load of each time slot can be expressed as: (11); In the formula, Indicates the first The basic load of each time slot Indicates the first The total power provided to all electric vehicles within each time slot.

[0053] The goal is to reduce the peak-valley load difference in microgrids, which can be expressed as: (12); In the formula, and These represent the maximum and minimum load requirements, respectively.

[0054] Step 4: Combining the constraints of fast and slow charging EVs and microgrids, the optimal charging time period is calculated using the CPLEX tool to provide data support for grid dispatch.

[0055] (1) The constraints for fast-charging electric vehicles are as follows: When the When an electric vehicle is identified as an emergency charging vehicle, its... Each arrival time slot is equal to 1. This ensures the vehicle continues charging from connection to the microgrid until it leaves, guaranteeing that emergency charging vehicles receive more power during charging periods. When the vehicle is disconnected from the microgrid, However, when the charging urgency of an electric vehicle is low, continuous use of fast charging power is not recommended. This could lead to overcharging. To prevent overcharging of fast-charging vehicles, constraints need to be set to limit their charging demand to no more than the maximum state of charge (SOC) demand. In conjunction with the above constraints, when a vehicle is connected to a microgrid and remains charging until it stops, The time slot for stopping charging is jointly determined by the off-grid time and the time slot for meeting the maximum SOC requirement. Constraints for emergency charging vehicles. It can be represented as (13); In the formula, The expression for the time slot indicating when a fast-charging vehicle stops charging is: (14); In the formula, To meet the maximum state of charge requirement, Indicates that the maximum is satisfied The required charging time slot This means always taking the largest integer less than the result of the division operation, thus preventing the electric vehicle from overcharging in the next integer time slot. This represents the upper limit of fast charging time calculated based on the maximum allowable battery loss. This represents the maximum loss factor for a single fast charge of the battery.

[0056] By accumulating the fast charging energy at different times, the loss coefficient of the battery capacity is ensured to be within the limit. To prevent excessive battery degradation, the cumulative energy limit constraints are as follows: (15); (2) Constraints for slow-charging electric vehicles are as follows: Due to the diverse charging needs of electric vehicle owners, each electric vehicle has a minimum and maximum SOC requirement to avoid overcharging. For slow-charging electric vehicles, the minimum SOC requirement must be met when the electric vehicle is disconnected from the microgrid, which can be expressed as: (16); In the formula, For the first The State of Charge (SOC) of an electric vehicle when disconnected from the microgrid can be calculated using the following formula: (17); In the formula, Represented as the first State of Charge (SOC) of an electric vehicle when connected to a microgrid.

[0057] The second constraint for slow-charging electric vehicles is related to scheduling time. Charging 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, its charging state... It must be equal to 0.

[0058] (3) Microgrid constraints The objective function is to reduce the peak-to-valley difference of the total load, but it cannot be guaranteed 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 the emergence of new charging peak loads in the microgrid during the coordinated scheme, the following constraints need to be imposed on the microgrid to limit the increase of the peak value.

[0059] (18); In the formula, This represents the maximum total load under a non-coordinated charging method that meets the maximum SOC demand of electric vehicles within 96 time slots.

[0060] Figure 3 The total load calculation process of the uncoordinated charging method of this invention is as follows: The system collects core charging parameters from electric vehicle users, including the charging start time, end time, initial state of charge (SOC) of the battery, and the user's preset minimum and maximum charge demand thresholds. Based on these parameters, the system performs charging behavior simulation, forcing all vehicles to charge continuously within a specified time window until the user-set demand threshold (minimum or maximum), generating an electric vehicle charging power curve. Subsequently, this curve is overlaid with the microgrid's basic load curve over time, and by accumulating the data time-by-time, the total load data for the public charging scenario is finally generated, providing a benchmark reference for subsequent optimized scheduling.

[0061] The static time-of-use electricity price ranges, prices, and EV parameter settings are shown in Tables 1 and 2, respectively: Table 1 Static Time-of-Use Electricity Price Ranges and Prices

[0062] Table 2 EV Parameter Settings

[0063] Description of scenarios for different scheduling methods: Scenario 1: Each mode initially has 50 vehicles, for a total of 100 vehicles. Without charging scheduling, the EVs are set to start charging immediately after connecting to the grid and continue charging until their charging needs are met or a predetermined departure time is reached, at which point charging stops.

[0064] Scenario 2: Based on the urgency index allocation model, users with high urgency will be fast-charged, while users with low urgency will be slow-charged. Taking into account the needs of both the grid side and the user side, the charging process of EVs will be rationally scheduled.

[0065] Scenario 3: Based on Scenario 2, and considering the different levels of importance attached to the urgency level at different times, the mode is adjusted according to the time period and urgency level of the vehicle.

[0066] Figure 4 The distribution of total microgrid load under different scheduling strategies is shown. The experimental results of three coordination scenarios intuitively reflect the core advantages of the optimization method of this invention in reducing peak-valley difference: In scenario 1, the total peak-valley difference is 462.09 kW, and the total electricity cost is 1829.02 yuan, indicating that the load fluctuation is significant under the initial strategy, and the system stability faces challenges. By reallocating the number of people charging in different modes, the total peak-valley difference is reduced to 429.36 kW, and the total electricity cost is 1587.37 yuan, with a peak-valley difference reduction of 7.1%, reflecting the smoothing effect of the dynamic urgency assessment model on the load curve. After further optimization, the total peak-valley difference is further reduced to 376.36 kW, and the total electricity cost is 1702.44 yuan, with a cumulative peak-valley difference reduction of 18.6%, verifying the synergistic effect of battery loss constraints and slow charging power allocation strategy in this invention, effectively alleviating the risk of grid overload and improving operating efficiency.

[0067] Figure 5 and Figure 6The distribution of slow charging and fast charging loads under different scheduling strategies is shown. Through data comparison across three scenarios, the optimization effect of the dynamic urgency assessment model of this invention on the load of each mode is intuitively demonstrated: In scenario 1, the peak-to-valley difference for slow charging is 115.50 kW, with an electricity cost of 837.98 yuan, while the peak-to-valley difference for fast charging is as high as 150.00 kW, with an electricity cost of 991.04 yuan. This indicates that under the initial fixed electricity price strategy, the fast charging load fluctuates drastically, leading to a significant risk of overload during peak hours. As the strategy is optimized to scenario 2, the peak-to-valley difference for slow charging remains at 115.50 kW, but the electricity cost rises to 1255.78 yuan, while the peak-to-valley difference for fast charging drops significantly to 80.00 kW, and the electricity cost decreases sharply to 331.59 yuan. By adjusting the allocation of 74 people for slow charging and 26 people for fast charging, this invention effectively suppresses the peak fast charging load, demonstrating the advantages of the dynamic threshold adjustment mechanism in hierarchical control of urgent needs. Further optimization to scenario 3 slightly increases the peak-to-valley difference for slow charging to 122.50 kW. The peak-valley difference in fast charging mode was further reduced to 20.00 kW and the electricity cost was 1505.12 yuan. The refined allocation strategy of 87 people for slow charging and 13 people for fast charging verified the synergistic effect of battery loss constraints and slow charging power spatiotemporal allocation model, successfully compressing the peak-valley difference of fast charging to 13.3% of the initial value, significantly alleviating the risk of equipment overload and improving charging efficiency.

[0068] This invention proposes an electric vehicle (EV) fast and slow charging scheduling optimization method based on dynamic urgency assessment. The method includes: establishing a dynamic selection model for EV charging modes based on charging urgency indicators; dividing the charging process into emergency fast charging and regular slow charging modes according to battery status, charging time windows, and user demand; and quantifying battery loss constraints during fast charging. For microgrid peak-valley regulation objectives, a multi-period optimization model is constructed, encompassing slow charging power allocation, fast charging priority scheduling, and microgrid safe operation, integrating charging demand stochastic simulation and load balancing constraints. A solution method combining Monte Carlo simulation and mixed-integer linear programming is employed to optimize the dynamic scheduling strategy in public charging scenarios. This invention effectively reduces the system peak-valley difference and the risk of fast charging equipment overload by differentiating charging demand levels and implementing refined time window management. Simultaneously, it meets the differentiated charging needs of most users, improving microgrid operational stability and renewable energy absorption capacity.

[0069] Example 2 In another preferred embodiment, based on Embodiment 1, this invention provides an electric vehicle fast and slow charging scheduling optimization method based on dynamic urgency assessment. Using a microgrid system of a public charging station in an office park as an application scenario, the effectiveness of the scheduling optimization method is verified through specific parameter configurations, experimental steps, and result analysis. This scenario involves the random access of 100 electric vehicles (EVs), 20 fast charging piles (60kW power) and 30 slow charging piles (7kW power). The microgrid's basic load is office power (including air conditioning, lighting, and office equipment). The scheduling cycle is 1 day (24 hours), and the time is discretized into 96 time slots (each time slot is 15 minutes, ΔT = 15 minutes).

[0070] I. Implementation Scenarios and Parameter Configuration 1.1 Basic Scene Setting Application scenario: A public charging station in a city's science and technology park serves the electric vehicles used by park employees for commuting. The charging behavior is concentrated between 8:00-9:00 am (when they arrive at work) and 5:00-6:00 pm (when they leave the site), which is consistent with the user behavior characteristics of public charging.

[0071] Equipment configuration: The microgrid includes one 1000kVA transformer. The base load is collected by the park's power distribution system, and the electric vehicle charging load is connected through the charging pile cluster. The dispatch center is equipped with a CPLEX 20.1 solver (for solving mixed integer optimization models).

[0072] 1.2 Definition of Core Parameters (1) Electric vehicle parameters (based on statistics of mainstream models in the market) are shown in Table 3: Table 3 Electric Vehicle Parameters

[0073] (2) User charging time parameters (normal distribution fitting) Based on one month's charging data statistics in the park, the arrival time of electric vehicles With departure time It follows the following normal distribution: Arrival time Expected value (8:30 AM), Standard Deviation (Fluctuation range 8:00-9:00), the distribution function is shown in formula (1); Departure time Expected value (5:30 PM), standard deviation (Fluctuation range 17:00-18:00), the distribution function is shown in formula (2).

[0074] The arrival / departure time is converted into a time slot number using the time discretization model [Formulas (3) and (4)], as shown in the following example: Arrival time of a certain vehicle ,but (Time slot 33, corresponding to 8:15-8:30); Departure time of a certain vehicle ,but (Time slot 70, corresponding to 17:30-17:45).

[0075] (3) Power grid time period division and dynamic threshold

[0076] Based on the load characteristics of the park's microgrid, a day is divided into three time periods, each corresponding to a different urgency threshold, as shown in Table 4: Table 4 Characteristics of the Park Microgrid

[0077] (4) Microgrid base load

[0078] Basic load data (unit: kW) was collected from 96 time slots through the park's power distribution system. The load during the core time period is shown in Table 5. Table 5 Load of the Park's Power Distribution System During Core Periods

[0079] II. Implementation Steps (Corresponding Invention Technical Solution) Step 1: Mathematical Modeling of Electric Vehicle Systems Based on the above parameters, three types of models are constructed: 1. User charging time model: The arrival times of 100 EVs are generated using a normal distribution function. With departure time Example: The first EV , ; 2. Time discretization model: The arrival / departure times of all EVs are converted into time slot numbers. The statistics show that the access time slot range of 100 EVs is 32~44 (peak hours) and the departure time slot range is 57~71 (off-peak hours). 3. Charging State Model: Initialize the charging state variables of all EVs. (Not charged), awaiting subsequent scheduling and allocation.

[0080] Step 2: Dynamic urgency assessment and mode determination of charging demand (1) Calculate the remaining charging period

[0081] Taking the first EV as an example, , ,but:

[0082] Corresponding charging time It meets the charging time requirements during office hours.

[0083] (2) Calculate the urgency index

[0084] Parameters of the first EV: , , , Substitute into formula (6):

[0085]

[0086]

[0087] (3) Determine the charging mode The first EV was connected during peak hours, specifically in slot 34. ,because Determined to be a non-urgent need, slow charging mode is assigned. ).

[0088] (4) Batch judgment results The CUI was calculated and the mode was determined for each of the 100 EVs. The results were statistically analyzed. Urgent demand (fast charging): 13 vehicles (CUI < corresponding time period threshold, mostly...) (Or access time slot is at the end of peak hours); Non-urgent demand (slow charging): 87 vehicles (CUI≥threshold for the corresponding time period, most of which had high SOC or sufficient charging time when connected).

[0089] Step 3: Load Overlay and Optimization Target Construction (1) Total load calculation Based on formula (11), the base load and charging load are superimposed, for example, time slot 33 (peak period, 8:15-8:30): base load ; Charging load: 13 fast-charging vehicles ( All vehicles are charging, and 87 vehicles are charging slowly (20 vehicles are charging after scheduling optimization). ); Total charging power: ; Total load: .

[0090] (2) Optimization Objective To minimize With the objective as the goal, an objective function (Formula 12) is constructed, and the constraints include fast and slow charging vehicle constraints and microgrid constraints.

[0091] Step 4: Solving for multiple constraints and generating scheduling instructions (1) Substituting the constraints Fast charging constraint: Taking the 5th EV (fast charging vehicle) as an example. Taking (e.g.) as an example, calculate the charging stop time slot. ) (Formula 14):

[0092]

[0093]

[0094] That is, the 5th EV will stop fast charging in the 36th time slot (9:00-9:15) to avoid overcharging and battery damage.

[0095] Microgrid constraints: Calculate the maximum total load for uncoordinated charging (charging all EVs upon connection). Peak total load after constraint optimization .

[0096] (2) CPLEX solution Input the parameters, constraints, and objective function of 100 EVs into CPLEX and solve for the following results: Solving time: 4.2 min (meets minute-level real-time scheduling requirements); Optimized total load peak Valley value Peak and valley power output is 376.36 kW.

[0097] III. Experimental Results and Effect Verification To verify the effectiveness of this invention, three sets of comparative experiments (scenario 1-3) were set up, and the comparison of the core indicators is shown in Table 6: Table 6 Comparison of Core Indicators

[0098] Key Effects Analysis 1. Improved grid stability: The total peak-to-valley difference of this invention (Scenario 3) is lower than that of uncoordinated charging (Scenario 1). This effectively smooths the load curve and mitigates the risk of overload during peak hours; 2. Equipment and battery safety: The peak-to-valley difference of fast charging has been reduced from 150kW to 20kW (compressed to the initial value of 13.3%), and the battery loss rate has been reduced from 8.2% to 2.1%, avoiding overcapacity of fast charging equipment and excessive battery loss; 3. User needs were met: All 13 vehicles with emergency needs were fully charged (SOC≥80%) before leaving, and the average SOC of the 87 slow-charging vehicles was 85% when disconnected, with a user satisfaction rate of 98%. 4. Economic balance: Although the total electricity cost is slightly higher than that of scenario 2 (due to the increased proportion of slow charging during off-peak hours), it is significantly lower than that of scenario 1, achieving a balance between "safety and economy".

[0099] This embodiment demonstrates, through specific parameter configuration and experimental verification in a public charging scenario within an office park, that the electric vehicle fast and slow charging scheduling optimization method based on dynamic urgency assessment is effective. 1. Accurately classify charging demand levels to adapt to the grid's carrying capacity at different times; 2. Quantify battery loss constraints to ensure safe operation of equipment and batteries; 3. Minimize the peak-to-valley difference in the microgrid and improve grid stability; 4. Enables large-scale scheduling at the minute level, adapting to the needs of real-world application scenarios.

[0100] This method can be directly extended to public charging scenarios such as commercial complexes and transportation hubs, providing practical technical support for the optimized scheduling of microgrids under the large-scale access of electric vehicles.

[0101] In the preferred embodiment, the charging data in step 1 includes the current battery level, charging time limit, minimum state of charge requirement, and maximum state of charge requirement. These settings ensure that the charging process accurately matches user needs, avoiding overcharging or undercharging. The system dynamically adjusts the charging power based on the minimum / maximum state of charge requirement and optimizes the charging curve in conjunction with the charging time limit, minimizing charging time while ensuring battery health.

[0102] In the preferred embodiment, the dynamic threshold adjustment rule in step 2 is as follows: during peak hours When it is determined to be an urgent need, during off-peak hours When it is determined to be an emergency need, during a stable period If a vehicle's access time slot is deemed an emergency, fast charging is activated; otherwise, slow charging is used. When a vehicle's access time slot spans multiple time periods, the period with the highest percentage of access is designated as the dominant period, and a corresponding threshold is applied. These settings ensure that the charging strategy flexibly adapts to different time-of-day demands: fast charging quickly replenishes energy during peak hours, slow charging protects the battery during off-peak hours, and selection is made as needed during stable periods. When a vehicle accesses the vehicle across different time periods, the dominant period threshold is used, effectively improving charging efficiency and resource utilization.

[0103] In the preferred embodiment, step 4 utilizes the parallel computing capabilities of the CPLEX tool to achieve real-time generation and dynamic tracking of large-scale vehicle group scheduling instructions. The CPLEX tool is used to solve the mixed-integer optimization model containing the constraints of the aforementioned formulas. These settings effectively shorten scheduling decision-making time and improve system response speed. Simultaneously, CPLEX's accurate solution capability ensures the optimality of the scheduling scheme, reduces vehicle empty mileage and waiting time, further lowers logistics and transportation costs, and improves overall operational efficiency.

[0104] In the preferred embodiment, the mathematical modeling in step 1 includes establishing a user charging time model. The arrival and departure times of electric vehicles follow a normal distribution, and the day is discretized into several 96 time slots, with each time slot having a length of 15 minutes. The time slot number represents the time when the vehicle connects to and disconnects from the microgrid. This setup can accurately depict the temporal distribution characteristics of the vehicle's charging and discharging behavior. Simultaneously, a charging demand power model is established. Based on battery capacity, initial charge, and target charge, combined with charging efficiency, the power required for each time slot is calculated, providing data support for subsequent microgrid power balance analysis.

[0105] In the preferred embodiment, the urgency index CUI in step 2 reflects the urgency of the electric vehicle's charging needs. This index characterizes whether slow charging can achieve the target SOC (State of Charge) during the vehicle's connection to the charging station. This setting accurately assesses the urgency of charging, avoiding the inability to meet travel needs on time due to blindly choosing slow charging. When the CUI value is high, it indicates that slow charging is unlikely to achieve the target SOC, and fast charging should be prioritized to ensure that the user's travel plans are not affected, thus improving the scientific nature of charging decisions.

[0106] In the preferred embodiment, when the time slot for a vehicle to access the microgrid spans multiple grid time periods in step 2, the time period with the largest proportion of the time period type during the access period is taken as the dominant time period, and the corresponding urgency threshold is adopted. The above settings can ensure that when a vehicle accesses the microgrid, the corresponding urgency threshold can be accurately matched according to the main characteristics of the actual time period, effectively avoiding judgment deviations caused by ambiguous time period divisions, and improving the accuracy and stability of the microgrid's response to vehicle access.

[0107] In the preferred scheme, the total microgrid load in step 3 includes base load, electric vehicle fast charging load, and slow charging load. The scheduling objective is achieved by minimizing the difference between the maximum and minimum demand of the total load. This setting effectively balances microgrid load fluctuations and improves system stability. Simultaneously, considering the differences in electricity prices at different times, the charging strategy is further optimized to reduce overall electricity costs while meeting user charging needs, achieving both economic and stability benefits.

[0108] In the preferred embodiment, the constraints on the fast-charging vehicle in step 4 include: the emergency charging vehicle continuously charges from connection to departure; the charging stop time slot is jointly determined by the off-grid time and the maximum SOC demand time slot; and the accumulated fast-charging energy does not exceed the limit calculated based on the maximum allowable battery loss. These settings ensure that the emergency charging vehicle can quickly replenish its power, effectively avoid damage to the battery caused by excessive fast charging, ensure that the battery operates within a safe and reasonable range, extend battery life, and improve the overall operating efficiency and stability of the charging system.

[0109] In the preferred embodiment, the constraints for slow-charging vehicles in step 4 include: the minimum SOC requirement must be met when the vehicle is disconnected from the microgrid, and charging can only be performed while the vehicle is connected to the microgrid. These settings ensure that slow-charging vehicles have sufficient range when disconnected from the grid, avoiding the impact of insufficient power on usage. At the same time, they standardize charging periods, prevent invalid charging operations in the non-connected state, and ensure the orderly operation of the microgrid and the rational allocation of resources.

[0110] In the preferred embodiment, the constraint condition for the microgrid in step 4 is: the total load peak after coordinated charging scheduling does not exceed the maximum total load value under the uncoordinated charging mode that meets the maximum SOC demand of electric vehicles within 96 time slots. This setting ensures that while meeting the charging needs of electric vehicles, grid overload caused by concentrated charging is effectively avoided. By dynamically adjusting the charging period, not only is grid stability improved, but charging costs are also reduced, achieving optimal resource allocation.

[0111] In summary, the electric vehicle fast and slow charging scheduling optimization method based on dynamic urgency assessment provided by this invention effectively solves the problem of reduced peak-valley regulation capacity caused by the spatiotemporal aggregation effect of charging load during the large-scale integration of electric vehicles into microgrids. Addressing the limitations of existing technologies, such as scheduling inaccuracies due to fuzzy hierarchical classification of emergency charging demand, equipment overload risks caused by the lack of battery loss constraints during fast and slow charging mode switching, difficulty in coordinating grid peak-valley regulation with differentiated user needs, and low efficiency in modeling the coupled random charging behavior and deterministic constraints of large-scale vehicle groups, this invention achieves breakthroughs through the following technical means: At the technical level, this invention proposes for the first time a scheduling optimization framework based on dynamic urgency assessment. By analyzing the remaining battery capacity, user charging time window, and grid operating status in real time, the charging mode is subdivided into emergency fast charging and regular slow charging, which solves the scheduling inaccuracy problem caused by the ambiguity of emergency demand classification in traditional methods. At the same time, a dynamic threshold adjustment mechanism is introduced to set the urgency index judgment threshold according to the characteristics of the grid peak, valley, and stable periods, so as to realize time-segmented charging optimization and control, breaking through the limitations of static threshold setting in the existing technology.

[0112] At the technical implementation level, this invention integrates quantitative constraints on battery loss with the safe operation boundary of a microgrid. In fast charging mode, it limits the maximum allowable battery loss, avoiding the risk of equipment operating beyond its capacity, and fills the gap in existing technologies for dynamic monitoring of battery health. Furthermore, by constructing a multi-time-period mixed-integer optimization model and combining it with Monte Carlo simulation to generate random charging demand scenarios, it achieves real-time generation and dynamic tracking of minute-level scheduling instructions, solving the problem of low efficiency in solving mixed-integer programming models for large-scale electric vehicle groups.

[0113] At the system optimization level, this invention designs a slow-charging power spatiotemporal allocation strategy, optimizing the regular slow-charging load to the grid's off-peak hours. Simultaneously, it prioritizes emergency vehicles for rapid full charging using urgency indicators, forming a three-level collaborative optimization framework of "demand hierarchy - power allocation - safety constraints," significantly improving the microgrid's operational stability. Furthermore, by using a minute-level time window discretization model to divide the day into 96 15-minute time slots, and combining this with the parallel computing capabilities of the CPLEX solver, it achieves minute-level response to large-scale vehicle group dispatching commands, overcoming the technical bottleneck of traditional static optimization models' lag in tracking the grid's basic load.

Claims

1.A method for electric vehicle fast and slow charging scheduling optimization based on dynamic urgency assessment, characterized in that, The method comprises the following steps: Step 1: Collect charging data of all electric vehicles and initialize a counter to process each vehicle; Step 2: Calculate the urgency index of each vehicle based on a charging urgency dynamic assessment model, dynamically adjust the threshold according to the peak, valley and stable periods of the power grid, determine the charging demand urgency, and directly enable fast charging for the urgent vehicles until the full charge state, and temporarily store the non-urgent vehicles until all vehicles are processed; Step 3: Superimpose the fast and slow charging loads of all vehicles on the basic grid load, and schedule based on the core target of minimizing the peak-valley load difference of the power grid; Step 4: Combine the constraints of fast and slow charging electric vehicles EV and micro-grid, calculate the optimal charging period through CPLEX tool, and provide data support for power grid scheduling. 2.The method of claim 1, wherein the method further comprises: The step 1 further comprises mathematical modeling of each part of the electric vehicle system, constructing an electric vehicle system containing user charging time window and time discretization in public scenarios, and the mathematical modeling comprises establishing a user charging time model and a time slot representation model. 3.The method of claim 2, wherein, The user charging time model and the time slot representation model are established by the following formula: Arrival time of an electric vehicle and departure time Subject to a normal distribution, the expression is: （1）； （2）； wherein and are the expected values of the public charging mode arrival and departure times, respectively; and are the standard deviations of the public charging mode arrival and departure times, respectively. A day is divided into 96 time slots, each time slot length The time slot number expression of the vehicle access and disconnection of the microgrid is: （3）； （4）； In the formula, Indicates the number of EVs; Indicates the first An index of electric vehicles; Indicates the first The time slot number for each electric vehicle to connect to the microgrid; Indicates the first The time slot number when an electric vehicle disconnects from the microgrid; Indicates the first The arrival time of the electric vehicle; Indicates the first The departure time of the electric vehicle; Indicates the length of a single time slot; Indicates the first Arrival time of electric vehicles Divide by the length of the time slot Round the result up. Indicates the first departure time of electric vehicles Divide by the length of the time slot The result is rounded down. 4.The method of claim 1, wherein the method further comprises: The charging data in the step 1 includes current power, charging time limit, minimum state of charge requirement and maximum state of charge requirement. 5.The method of claim 1, wherein, The specific implementation of the step 2 comprises the following formula and rules: (1) The number of remaining time periods of the vehicle connected to the micro-grid: （5）； (2) Urgency indicator : （6）； (3) Charging mode determination: （7）； Wherein, （8）； (4) Charging power: （9）； (5) Charging state: （10）； In the formula, For the first The remaining number of time slots during which electric vehicles can continue to be connected to the microgrid; For charging efficiency; Battery capacity; , These are the fast charging and slow charging powers, respectively. Indicates the first The vehicle is connected to the charging station , For the first When the vehicle finishes charging The lower bound; The time period in which the vehicle is located; This indicates whether the charging needs of each EV are urgent; For time-period dynamic threshold parameters; Indicates the first The car in The charging status over a period of time. 6.The method of claim 1, wherein, The dynamic threshold adjustment rule in step 2 is: high peak period Emergency demand is determined when the low valley period Emergency demand is determined when the smooth period Emergency demand is determined when the fast charging is enabled, otherwise slow charging is adopted; when the vehicle access time slot spans multiple periods, the period with the largest proportion is taken as the dominant period and the corresponding threshold is adopted. 7.The method of claim 1, wherein, The load superposition and scheduling target in the step 3 are realized by the following formula: (1) Total load: （11）； (2) Scheduling target: （12）； wherein is the total load of the th time slot; is the basic load of the th time slot; denotes the charging power of the th vehicle; , are the maximum and minimum values of the total load, respectively. 8.The method of claim 1, wherein, The constraint conditions of fast charging vehicles in the step 4 comprise the following formula: (1) State of charge constraint : （13）； (2) Stop charging time slot: （14）； (3) Cumulative energy constraint: （15）； In the formula, is the fast charging stop time slot; is the maximum loss coefficient of the battery single fast charging. 9.The method of claim 1, wherein, The constraint conditions of slow charging vehicles in the step 4 comprise the following formula: (1) State of charge constraint: （16）； (2) Disconnected state of charge: （17）； (3) charging time constraint: when the vehicle is not connected to the microgrid, ; In the formula, SOCoff is the state of charge when the vehicle is disconnected from the microgrid. 10.The method of claim 1, wherein, The constraint conditions of the micro-grid in the step 4 are as follows: （18）； In the formula, is the maximum value of the total load in the non-coordinated charging mode that meets the maximum SOC demand of the electric vehicle in 96 time slots.

Images