A method for quantitatively evaluating multi-element flexibility of an electric vehicle cluster

Abstract

The application discloses a kind of electric vehicle cluster multi-element flexibility quantitative evaluation method, multi-element flexibility modeling is carried out for electric vehicle cluster, the operation boundary model of EV cluster is obtained, and according to the flexibility constraint of EV cluster, with the minimum generation cost of electric energy market as optimization goal, the joint clearing model of power system energy and auxiliary service market participated by EV cluster is proposed, and flexibility penalty term is added in joint clearing model, to be used for quantifying the multi-element flexibility resource of EV cluster;In each random scene simulated, the optimized joint clearing model is solved, and the quantitative evaluation of the size of the multi-element flexibility potential provided by EV cluster is realized according to the data results of simulation solution, and then the data basis for the related research of the multi-element flexibility of EV cluster participating in electric power market is provided.

Description

Technical Field

[0001] This invention relates to the field of power system flexibility technology, and in particular to a method for quantitatively evaluating the multi-dimensional flexibility of electric vehicle clusters. Background Technology

[0002] Under the "dual carbon" context, the power industry is gradually undertaking energy transformation, vigorously developing new energy sources, primarily wind and solar power. While the high proportion of new energy connected to the grid brings benefits such as cleanliness, efficiency, and low carbon emissions, its highly volatile and uncertain characteristics also pose new challenges to the safe and stable operation of the power system. The model of relying solely on generation-side flexibility is no longer sufficient to meet the flexibility requirements of the new power system. Electric vehicles, as a rapidly growing emerging load, have a large adjustability and good flexibility potential; therefore, research on the flexibility of electric vehicle clusters is of great significance.

[0003] Scholars have already conducted research on the flexibility of electric vehicles (EVs), with current research directions mainly falling into two categories. One category studies Vehicle-to-Grid (V2G) technology, proposing control strategies for EV clusters to provide frequency regulation and other ancillary services. The other category studies the charging behavior of EV users, establishing mathematical models for EV clusters to participate in grid optimization scheduling, and proposing optimal scheduling schemes or market strategies for EV clusters to provide flexibility. To better utilize the flexibility potential of EV clusters, quantitative analysis of their flexibility levels is needed. However, due to the complexity of EV cluster charging behavior, quantitative assessment of EV cluster flexibility is difficult, and there is a lack of quantitative research methods for providing diverse flexibility capacities. This results in a lack of data support when EV clusters participate in the electricity market to provide flexibility, making it difficult to quantify their costs and benefits. Summary of the Invention

[0004] This invention provides a method for quantitatively evaluating the multi-dimensional flexibility of electric vehicle clusters to overcome the aforementioned technical problems.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows:

[0006] A method for quantitatively evaluating the multi-dimensional flexibility of electric vehicle clusters includes the following steps:

[0007] S1: Obtain the charging parameters of the EV charging station and the EV operating parameters to establish a mathematical model of the operating boundary of a single EV.

[0008] The charging parameters of the EV charging station include at least the maximum charging power, maximum discharging power, and number of charging piles under different charging modes.

[0009] The EV operating parameters include at least the maximum / minimum permissible state of charge of the battery and the charging time;

[0010] S2: Obtain historical data on different types of EV charging at EV charging stations, and fit the probability distribution parameters of EV charging patterns to simulate and generate multiple random scenarios;

[0011] Based on the simulated random scenarios, the mathematical model of the single EV's operating boundary is aggregated using the Minkowski summation method to obtain the operating boundary model of the EV cluster.

[0012] S3: Establish the flexibility constraints of the EV cluster based on the operational boundary model of the EV cluster; and based on the flexibility constraints of the EV cluster, construct a joint clearing model for the power system energy and ancillary services market in which the EV cluster participates, with the lowest power generation cost in the electricity market as the optimization objective.

[0013] Furthermore, based on the joint clearing model, cost reduction items brought about by the flexibility provided by the EV cluster are added, and the joint clearing model is optimized to quantify the diverse flexibility resource potential of the EV cluster.

[0014] S4: Based on the simulation and solution of various random scenarios, the optimized joint clearing model is solved, and the data results of the simulation and solution are used to realize the quantitative assessment of the multiple flexibility provided by EV clusters in participating in the electricity market.

[0015] Furthermore, the mathematical model of the single EV operating boundary established in S1 includes the time-series energy boundary model and the time-series power boundary model of a single EV.

[0016] And the expression for the time-series energy boundary model of a single EV is:

[0017]

[0018] In the formula: E i，t These represent the upper and lower bounds of the energy of a single EV, respectively; E i,0 , These represent the initial battery level of the EV when it enters the charging station and the expected battery level when it leaves; These represent the upper limit of the achievable capacity of the EV battery and the lower limit of the capacity that must be retained, respectively. These represent the maximum charging power and maximum discharging power of the EV, respectively. These represent the time periods when the EV enters and leaves the charging station, respectively; Δt represents the step size of a single time period; t represents the current time period;

[0019] The expression for the time-series power boundary model of a single EV is:

[0020]

[0021] In the formula: P i,t These are represented as the upper and lower bounds of the energy EV, respectively. E i,t+1 These represent the upper and lower bounds of the energy of a single EV during time period t+1, respectively.

[0022] Furthermore, S2 includes the following steps:

[0023] S21: Obtain historical charging data of EV cluster users from EV charging stations to obtain daily charging pattern characteristics of EV cluster users, and classify EV cluster users according to the charging pattern characteristics.

[0024] The charging pattern characteristics include start charging time, end charging time, initial charge level, and expected charge level; and the different types of EV users include at least private electric vehicles, ride-hailing electric vehicles, and electric buses;

[0025] The probability distribution of EV charging pattern characteristics was obtained by fitting the data using Python.

[0026] S22: Using the Monte Carlo simulation method, multiple random charging scenarios for EV charging stations at different times within a single day are generated based on the probability distribution of the EV charging pattern.

[0027] S23: Based on the simulated random scenario, the mathematical model of the single EV operating boundary is aggregated using the Minkowski summation method to obtain the operating boundary model of the EV cluster; and the operating boundary model of the EV cluster includes the time-series energy boundary model and the time-series power boundary model of the EV cluster.

[0028] The expression for the time-series energy boundary model of the EV cluster is:

[0029]

[0030] In the formula: B represents the upper and lower bounds of the energy for the EV cluster, respectively; i,t N represents the state variable indicating whether the i-th EV is on the network during time period t; EV This represents the maximum number of EVs within the EV cluster;

[0031] The mathematical expression for the timing power boundary of the EV cluster is:

[0032]

[0033] In the formula: These represent the upper and lower power bounds of the EV cluster, respectively.

[0034] Furthermore, S3 includes the following steps:

[0035] S31: Considering the energy and power requirements of the power system's multi-dimensional flexibility at different time scales, establish the flexibility constraints of the EV cluster based on the operational boundary model of the EV cluster.

[0036] The aforementioned multi-faceted flexibility refers to the power system's ability to economically utilize various flexible resources to cope with uncertainties in power sources, power grids, and loads in order to maintain a dynamic balance between power supply and demand.

[0037] Furthermore, the response time requirement for the flexibility of the EV cluster is reflected by the upper limit constraint of the auxiliary services, and the expression for the upper limit constraint of the auxiliary services is as follows:

[0038]

[0039] In the formula: These represent the capacity of the EV cluster to provide upward and downward type k auxiliary services to the EV cluster by adjusting the charging and discharging power speeds, respectively. These represent the upper limits of the k-th type of auxiliary services provided by the charging and discharging power adjustment speed of the EV cluster to the EV cluster, respectively.

[0040] The duration requirement for EV cluster flexibility is reflected by the capacity stacking constraint and power accumulation constraint of the ancillary services. The expression for the capacity stacking constraint is:

[0041]

[0042] In the formula: This represents the capacity of the EV cluster to provide the k-th type of upward auxiliary service during the time period t-1; This represents the capacity of the EV cluster to provide the k-th type of downward auxiliary service during the time period t-1; This indicates the lower bound of the power of the EV cluster; This represents the upper limit of the power of the EV cluster; These represent the charging and discharging power of the EV cluster, respectively; k represents the type of ancillary service; A represents the set of ancillary service types with a duration requirement of less than one time period; B represents the set of ancillary service types with a duration requirement of more than one time period.

[0043] The power constraint takes into account the cumulative effect of providing ancillary services, and the power accumulation constraint is specifically as follows:

[0044] When k∈A, the expression for the energy accumulation constraint is:

[0045]

[0046] In the formula: Δt represents the step size of a single time period; Indicates the lower bound of the energy of the EV cluster; E represents the upper bound of the energy of the EV cluster; t Represented as the battery level of the EV cluster; t k Let t represent the duration of the ancillary service requirement for class k, and t k ≤1; η EV,ch ,η EV,dis These represent the charging and discharging efficiencies of the EV cluster, respectively.

[0047] When k∈B, the expression for the energy accumulation constraint is:

[0048]

[0049] In the formula: t k Let t represent the duration of the ancillary service requirement for class k, and t k ≤1.

[0050] S32: Based on the flexibility constraints of the EV cluster, and with the goal of minimizing the generation cost of the power market, construct a joint clearing model for the power system energy and ancillary services market in which the EV cluster participates.

[0051] Furthermore, the joint clearing model includes an objective function and constraints.

[0052] The constraints include at least the operating constraints of thermal power units, system power balance constraints, and the flexibility constraints of EV clusters.

[0053] It also adds cost reduction items generated by the flexibility resources provided by EV clusters, and optimizes the joint clearing model as a multi-dimensional flexibility indicator for quantifying EV clusters.

[0054] The optimized expression for the joint clearing model is as follows:

[0055]

[0056]

[0057] In the formula: min C represents the objective function; This represents the power generation cost of thermal power unit i during time period t; This represents the startup status variable of the generator unit; V represents the unit's shutdown status variable; V represents the total number of time periods. Represented as the coefficients of a quadratic function of the power generation cost of thermal power units; These represent the single start-up and shutdown costs of thermal power units, respectively; n g The number of thermal power units is represented by M; M represents the cost reduction items generated by flexibility resources; a k Represented as the cost compensation coefficient for type k flexible resources; This represents the demand capacity of flexible resources of type k during time period t; This represents the total capacity of the EV cluster to provide k types of flexible resources during time period t.

[0058] Furthermore, S4 includes the following steps:

[0059] S41: In each random charging scenario, by calling the CPLEX optimization toolbox in Matlab, the optimized joint clearing model is optimized and simulated to solve based on the constraints, and the actual power / energy and available multi-dimensional flexibility capacity data of the EV cluster at different time periods within a single day are obtained in the corresponding random charging scenario.

[0060] S42: Based on the multivariate flexibility capacity data provided by the EV cluster in the corresponding random charging scenario, the mean value of the probability distribution function of the flexibility capacity of the EV cluster at different times of the day is obtained by fitting the data using Python's fitter package.

[0061] S43: The mean of the probability distribution function of the flexibility capacity of the EV cluster at different times within a single day is selected as the evaluation index of the flexibility potential, so as to realize the quantitative evaluation of the EV cluster's participation in the power market at different times within a single day to provide diversified flexibility.

[0062] Beneficial Effects: This invention provides a method for quantitatively evaluating the multi-faceted flexibility of electric vehicle (EV) clusters. It models the multi-faceted flexibility of EV clusters, obtains the operational boundary model of the EV cluster, and proposes a joint clearing model for the power system energy and ancillary services market in which EV clusters participate, based on the flexibility constraints of the EV cluster and with the minimum power generation cost in the electricity market as the optimization objective. A flexibility penalty term is added to the joint clearing model to quantify the multi-faceted flexibility resource potential of the EV cluster. The optimized joint clearing model is solved under various simulated stochastic scenarios, and the data results from the simulations are used to quantitatively evaluate the multi-faceted flexibility that EV clusters can provide. This provides a data foundation for related research on the multi-faceted flexibility provided by EV clusters participating in the electricity market. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 This is a flowchart of a method for quantitatively evaluating the multi-dimensional flexibility of electric vehicle clusters according to the present invention.

[0065] Figure 2 This is a flowchart illustrating the method for quantifying the multi-dimensional flexibility of electric vehicle clusters in this embodiment.

[0066] Figure 3 This is a schematic diagram of the improved standard 6-node system in this embodiment;

[0067] Figure 4 This is a schematic diagram of the power and energy boundaries of an EV cluster under a random charging scenario in this embodiment;

[0068] Figure 5 This is a simulation diagram showing the quantification results of four types of flexibility resources provided by the EV cluster in a random charging scenario in this embodiment.

[0069] Figure 6 This is a Monte Carlo simulation result of the EV cluster's upward secondary frequency modulation capacity in this embodiment;

[0070] Figure 7 This is a schematic diagram of the secondary frequency modulation capacity distribution fitting of the EV cluster at 22:00 in this embodiment;

[0071] Figure 8 This is a schematic diagram illustrating the quantification of the upward secondary frequency modulation flexibility level of the EV cluster in this embodiment. Detailed Implementation

[0072] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0073] This embodiment provides a method for quantitatively evaluating the multi-dimensional flexibility of electric vehicle clusters, such as... Figures 1 to 2 As shown, it includes the following steps:

[0074] S1: Obtain the charging parameters of the EV charging station and the EV operating parameters to establish a mathematical model of the operating boundary of a single EV.

[0075] The charging parameters of the EV charging station include at least the maximum charging power, maximum discharging power, and number of charging piles under different charging modes.

[0076] The EV operating parameters include at least the maximum / minimum permissible state of charge of the battery and the charging time;

[0077] Specifically, EV charging stations offer various charging modes, the main difference being the upper limit of the charging power. When establishing the mathematical model for the operational boundary of a single EV, charging modes are divided into fast charging and slow charging. Fast charging offers high charging power and a short EV stay on the grid; slow charging offers low charging power and a long EV stay. Once an EV enters a charging station and begins charging, its upper energy limit is: charging at maximum power until the upper limit is reached and then maintained. The lower energy limit is: ensuring the EV reaches the user's desired minimum charge level upon leaving the grid, discharging at maximum power to the lower limit before recharging. If the EV's dwell time is less than or equal to a preset scheduling period, there is no discharge period, and its lower energy limit is: charging at a lower power, reaching the owner's desired minimum charge level upon departure. The upper and lower energy limits of the EV can be used to determine the schedulable area of ​​the EV in a time-series energy graph. Within the geometric space formed by the upper and lower energy limits, any broken line connecting the start and departure times represents a feasible charging and discharging scheme. The upper limit of EV charging and discharging power is the smaller of the maximum change in battery power between adjacent time periods and the maximum charging and discharging power of the charging station.

[0078] The established mathematical model for the operating boundary of a single EV includes a time-series energy boundary model and a time-series power boundary model for a single EV.

[0079] And the expression for the time-series energy boundary model of a single EV is:

[0080]

[0081] In the formula: E i，t These represent the upper and lower bounds of the energy of a single EV, respectively; E i,0 , These represent the initial battery level of the EV when it enters the charging station and the expected battery level when it leaves; These represent the upper limit of the achievable capacity of the EV battery and the lower limit of the capacity that must be retained, respectively. These represent the maximum charging power and maximum discharging power of the EV, respectively. These represent the time periods when the EV enters and leaves the charging station, respectively; Δt represents the step size of a single time period; t represents the current time period;

[0082] The expression for the time-series power boundary model of a single EV is:

[0083]

[0084] In the formula: P i,t These are represented as the upper and lower bounds of the energy EV, respectively. E i,t+1 These represent the upper and lower bounds of the energy of a single EV during time period t+1, respectively.

[0085] S2: Obtain historical data on different types of EV charging at EV charging stations, and fit the probability distribution parameters of EV charging patterns to simulate and generate multiple random scenarios;

[0086] Based on the simulated random scenarios, the mathematical model of the single EV's operating boundary is aggregated using the Minkowski summation method to obtain the operating boundary model of the EV cluster.

[0087] Specifically, the following steps are included:

[0088] S21: Obtain historical charging data of EV cluster users from EV charging stations to obtain daily charging pattern characteristics of EV cluster users, and classify EV cluster users according to the charging pattern characteristics.

[0089] The charging pattern characteristics include start charging time, end charging time, initial charge level, and expected charge level; and the different types of EV users include at least private electric vehicles, ride-hailing electric vehicles, and electric buses;

[0090] Specifically, different types of EVs exhibit different charging patterns. For instance, private car charging typically occurs during commuting hours and evening rest periods, while ride-hailing users charge during off-peak hours when electricity prices are lower. Electric buses tend to use slower charging methods, which are cheaper and result in longer charging times, often concentrated at night. Historical data on the start charging time, departure time, initial charge level, and expected charge level for different types of EVs were fitted.

[0091] The probability distribution of EV charging pattern characteristics was obtained by fitting the data using Python.

[0092] S22: Based on the probability distribution of EV charging pattern characteristics, charging scenarios of EV charging stations within a day can be randomly generated. Using the Monte Carlo simulation method, multiple random charging scenarios of EV charging stations at different time periods within a single day are obtained based on the probability distribution of the EV charging pattern, which improves the accuracy of the flexibility evaluation results.

[0093] S23: Based on the simulated random scenario, the mathematical model of the single EV operating boundary is aggregated using the Minkowski summation method to obtain the operating boundary model of the EV cluster.

[0094] Furthermore, the operational boundary model of the EV cluster includes the time-series energy boundary model and the time-series power boundary model of the EV cluster;

[0095] The expression for the time-series energy boundary model of the EV cluster is:

[0096]

[0097] In the formula: B represents the upper and lower bounds of the energy for the EV cluster, respectively; i,t N represents the state variable indicating whether the i-th EV is on the network during time period t; EV This represents the maximum number of EVs within the EV cluster;

[0098] The mathematical expression for the timing power boundary of the EV cluster is:

[0099]

[0100] In the formula: These represent the upper and lower power bounds of the EV cluster, respectively.

[0101] S3: Establish the flexibility constraints of the EV cluster based on the operational boundary model of the EV cluster; and based on the flexibility constraints of the EV cluster, construct a joint clearing model for the power system energy and ancillary services market in which the EV cluster participates, with the lowest power generation cost in the electricity market as the optimization objective.

[0102] Furthermore, based on the joint clearing model, cost reduction items brought about by the flexibility provided by the EV cluster are added, and the joint clearing model is optimized to quantify the diverse flexibility resource potential of the EV cluster.

[0103] Specifically, the following steps are included:

[0104] S31: Considering the energy and power requirements of the power system's multi-dimensional flexibility at different time scales, establish the flexibility constraints of the EV cluster based on the operational boundary model of the EV cluster.

[0105] The aforementioned multi-faceted flexibility refers to the power system's ability to economically utilize various flexible resources to cope with uncertainties in power sources, power grids, and loads within a preset timeframe in order to maintain a dynamic balance between power supply and demand.

[0106] Referring to existing flexibility classification methods at home and abroad, the multiple flexibility can be summarized and classified at different time scales: namely, primary frequency regulation capacity, secondary frequency regulation capacity, 10-minute standby capacity and 30-minute standby capacity, and the duration requirements of the four types of ancillary services are 1min, 10min, 30min and 2h respectively.

[0107] Furthermore, the response time requirement for the flexibility of the EV cluster is reflected by the upper limit constraint of the auxiliary services, and the expression for the upper limit constraint of the auxiliary services is as follows:

[0108]

[0109] In the formula: These represent the capacity of the EV cluster to provide upward and downward type k auxiliary services to the EV cluster by adjusting the charging and discharging power speeds, respectively. These represent the upper limits of the k-th type of auxiliary services provided by the charging and discharging power adjustment speed of the EV cluster to the EV cluster, respectively.

[0110] The duration requirement for EV cluster flexibility is reflected by the capacity stacking constraint and power accumulation constraint of the ancillary services. The expression for the capacity stacking constraint is:

[0111]

[0112] In the formula: This represents the capacity of the EV cluster to provide the k-th type of upward auxiliary service during the time period t-1; This represents the capacity of the EV cluster to provide the k-th type of downward auxiliary service during the time period t-1; This indicates the lower bound of the power of the EV cluster; This represents the upper limit of the power of the EV cluster; These represent the charging and discharging power of the EV cluster, respectively; k represents the type of ancillary service; A represents the set of ancillary service types with a duration requirement of less than one time period; B represents the set of ancillary service types with a duration requirement of more than one time period.

[0113] Furthermore, if the time scale for flexibility is set to 1 hour as the single time interval, the expression for the capacity superposition constraint can be rewritten as follows:

[0114]

[0115]

[0116] In the formula: when k takes the value of 1, 2, 3, and 4, it represents the primary frequency modulation capacity, the secondary frequency modulation capacity, the 10-minute standby capacity, and the 30-minute standby capacity, respectively.

[0117] The power constraint takes into account the cumulative effect of providing ancillary services, and the power accumulation constraint is specifically as follows:

[0118] When k∈A, the expression for the energy accumulation constraint is:

[0119]

[0120] In the formula: Δt represents the step size of a single time period; Indicates the lower bound of the energy of the EV cluster; E represents the upper bound of the energy of the EV cluster; t Represented as the battery level of the EV cluster; t k Let t represent the duration of the ancillary service requirement for class k, and t k≤1; η EV,ch ,η EV,dis These represent the charging and discharging efficiencies of the EV cluster, respectively.

[0121] Furthermore, if the time scale for flexibility is set to 1 hour as the single time interval, the energy accumulation constraint is rewritten as follows:

[0122]

[0123] When k∈B, the expression for the energy accumulation constraint is:

[0124]

[0125] In the formula: t k Let t represent the duration of the ancillary service requirement for class k, and t k ≤1.

[0126] Furthermore, if the time scale for flexibility is set to 1 hour as the single time interval, the energy accumulation constraint is rewritten as follows:

[0127]

[0128] S32: Based on the flexibility constraints of the EV cluster, and with the goal of minimizing the generation cost of the power market, construct a joint clearing model for the power system energy and ancillary services market in which the EV cluster participates.

[0129] Furthermore, the joint clearing model includes an objective function and constraints.

[0130] The constraints include at least the operating constraints of thermal power units, the power balance constraints of the power system, and the flexibility constraints of EV clusters.

[0131] The flexibility constraints of the EV cluster include upper limit constraints for ancillary services, capacity aggregation constraints for ancillary services, and power accumulation constraints for ancillary services.

[0132] The objective function includes generation cost and start-up / shutdown cost, expressed as:

[0133]

[0134] The operating constraints of the thermal power unit include at least the output constraints of the thermal power unit and the minimum start / stop time constraints.

[0135] Thermal power unit output constraints:

[0136]

[0137] In the formula: The power output of the thermal power unit during time period t; These are start and stop state variables; These are the upper and lower limits of the output of thermal power units, respectively.

[0138] Minimum start / stop time constraints for thermal power units:

[0139]

[0140]

[0141]

[0142] In the formula: x g,i,t y g,i,t These represent the start-up and shutdown status variables of the unit, respectively; U i D i These represent the minimum start-up and shutdown times of the unit, respectively.

[0143] Power system power balance constraints:

[0144]

[0145] In the formula, D t This represents the total load of the system during time period t.

[0146] The constraints on the EV cluster also include power constraints, charge / discharge mutual exclusion constraints, upper and lower limit constraints on battery capacity, and battery balance constraints. Power constraints ensure that the actual charging and discharging power of the EV cluster remains within power boundaries. Charge / discharge mutual exclusion constraints ensure that the EV cluster will not charge and discharge simultaneously. Upper and lower limit constraints ensure that the actual operating battery capacity of the EV cluster remains within battery capacity boundaries. Battery balance constraints indicate that the charge and discharge amounts of the EV cluster should be equal throughout the day. The above constraints are as follows:

[0147] Power constraints:

[0148]

[0149] In the formula: These are the charging and discharging state variables of the EV cluster, respectively.

[0150] Charge and discharge mutual exclusion constraint:

[0151]

[0152] Battery level upper and lower limits constraints:

[0153]

[0154] Power balance constraints:

[0155]

[0156] In the formula: E t-1This represents the battery level of the EV cluster during time period t-1;

[0157] By adding cost reduction items generated by the flexibility resources provided by EV clusters, the joint clearing model is optimized as a way to quantify the multi-faceted flexibility of EV clusters.

[0158] The optimized expression for the joint clearing model is as follows:

[0159]

[0160]

[0161] In the formula: minC represents the objective function; This represents the power generation cost of thermal power unit i during time period t; This represents the startup status variable of the generator unit; V represents the unit's shutdown status variable; V represents the total number of time periods. Represented as the coefficients of a quadratic function of the power generation cost of thermal power units; These represent the single start-up and shutdown costs of thermal power units, respectively; n g The number of thermal power units is represented by M; M represents the cost reduction items generated by flexibility resources; a k Represented as the cost compensation coefficient for type k flexible resources; This represents the demand capacity of flexible resources of type k during time period t; This represents the total capacity of the EV cluster to provide k types of flexible resources during time period t.

[0162] S4: Based on the simulation and solution of various random scenarios, the optimized joint clearing model is solved, and the data results of the simulation and solution are used to realize the quantitative assessment of the multiple flexibility provided by EV clusters in participating in the electricity market.

[0163] Specifically, the following steps are included:

[0164] S41: In each random charging scenario, by calling the CPLEX optimization toolbox in Matlab, the optimized joint clearing model is optimized and simulated to solve based on the constraints, and the actual power / energy and available multi-dimensional flexibility capacity data of the EV cluster at different time periods within a single day are obtained in the corresponding random charging scenario.

[0165] Furthermore, the available diverse flexibility capacity data is determined by the actual power / energy of the EV cluster and the constraints.

[0166] S42: Based on the multivariate flexibility capacity data provided by the EV cluster in the corresponding random charging scenario, the mean value of the probability distribution function of the flexibility capacity of the EV cluster at different times of the day is obtained by fitting the data using Python's fitter package.

[0167] S43: The mean of the probability distribution function of the flexibility capacity of the EV cluster at different times within a single day is selected as the evaluation index of the flexibility potential, so as to realize the quantitative evaluation of the EV cluster's participation in the power market at different times within a single day to provide diversified flexibility.

[0168] In this embodiment, an improved 6-node system is used as a simulation example for analysis, such as... Figure 3 As shown, the improved 6-node system includes 3 thermal power units (corresponding to nodes 1, 2, and 6 respectively) and 1 wind power unit (corresponding to node 4). The EV charging station cluster is connected to node 4. The 6-node system has loads connected to nodes 3, 4, 5, and 6. The first transformer T1 is located between nodes 2 and 3, and the second transformer T2 is located between nodes 4 and 5. Three common EV types are considered: buses, private cars, and ride-hailing vehicles. Buses use slow charging, while private cars and ride-hailing vehicles use fast charging. Simulation information on charging characteristics is shown in Table 1. The total number of EVs in the three types is 1000, 3000, and 1500 respectively; the individual battery capacities are 150kW, 32kW, and 50kW respectively. The maximum charging and discharging power of fast charging is 30kW, and the maximum charging and discharging power of slow charging is 10kW.

[0169] Table 1 Simulation information on charging status of various electric vehicles

[0170]

[0171] Based on the simulation information of charging pattern characteristics, Monte Carlo simulation was carried out to generate random charging scenarios. By calling the CPLEX optimization toolbox in Matlab, the established joint clearing model was optimized and solved to obtain the actual power / energy and the available multi-flexibility capacity data of EV clusters at different time periods within a single day under the corresponding random charging scenario.

[0172] Based on the multi-dimensional flexibility capacity data of EV clusters obtained from simulations under 1000 randomly generated scenarios, the probability distribution of multi-dimensional flexibility capacity provided by EV clusters at different time periods of a day is solved by data fitting algorithm; and the mean of the probability distribution of multi-dimensional flexibility capacity provided by EV clusters at different time periods of a day is selected to achieve a quantitative assessment of the flexibility provided by EV clusters in participating in the electricity market.

[0173] Figure 4The diagram shows the operational boundaries of the EV cluster. The dashed lines represent the upper and lower boundaries of power and energy, while the solid lines represent the actual power and energy operating curves. The space between the solid and dashed lines indicates the potential flexibility of the EV cluster. Due to user usage patterns, the number of EVs charging on the grid peaks at night (8:00 PM to 5:00 AM the next day). Therefore, the variable range of power and energy is larger at night, indicating greater flexibility potential.

[0174] like Figure 5 The diagram shows the capacity that an EV cluster can simultaneously provide for four types of flexible resources. When providing all four ancillary services, the capacity available for secondary frequency regulation and 10-minute standby is greater. The capacity of primary frequency regulation is limited by the controller droop factor; while 30-minute standby requires a long duration and has a coupling relationship between adjacent time periods, requiring a large amount of power space, so the EV cluster can only provide a relatively small 30-minute standby capacity. To maximize the supply of flexibility, a larger portion of the available flexible capacity is allocated to primary frequency regulation, secondary frequency regulation, and 10-minute standby. This aligns with the physical characteristics of load-side flexible resources, which are better suited for providing flexibility resources that require rapid response and short durations.

[0175] Meanwhile, Monte Carlo simulations were further conducted to generate 1,000 random scenarios and analyze the optimization results. Figure 6 The Monte Carlo simulation results of the EV cluster's upward secondary frequency modulation capacity are presented. Based on the calculation results under a large number of random scenarios, the trend and overall level of the EV cluster's upward secondary frequency modulation capacity can be predicted within a day. The available secondary frequency modulation capacity is highest from 22:00 to 5:00 the next day; the secondary frequency modulation capacity level is second highest from 16:00 to 20:00; and the available secondary frequency modulation capacity is lower during other time periods.

[0176] Taking the 22:00 time period as an example, a fitting analysis is performed on the quantization results of the secondary frequency modulation. For example... Figure 7 The image shows a histogram of capacity distribution under random scenarios and several well-fitting curves. The gamma distribution has the smallest fitting error, and its expected value (16618.89 kW) can be calculated from the fitting parameters. Data from 24 time periods are fitted separately, and the expected value of the optimal distribution function is returned as the evaluation result. Figure 8 As shown, this is the assessment result of the flexibility level of the EV cluster's two frequency adjustments within a day.

[0177] This embodiment establishes a market clearing model that considers multiple flexibility factors by modeling and aggregating the operational boundaries of EV clusters. Specifically, it adds a cost reduction term for EV-provided flexibility and employs Monte Carlo simulation to evaluate the flexibility level. Compared to traditional optimization models for EV participation in grid dispatch, this flexibility quantification method considers the time scale requirements of flexibility, enabling the allocation and quantification of flexibility potential among multiple flexibility factors based on the physical characteristics of EV clusters. Furthermore, it considers the uncertainty of EV user charging patterns when evaluating flexibility levels, thus improving accuracy.

[0178] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention 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 or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for quantitatively evaluating the multi-dimensional flexibility of electric vehicle clusters, characterized in that, Includes the following steps: S1: Obtain the charging parameters of the EV charging station and the EV operating parameters to establish a mathematical model of the operating boundary of a single EV. The charging parameters of the EV charging station include at least the maximum charging power, maximum discharging power, and number of charging piles under different charging modes. The EV operating parameters include at least the maximum / minimum permissible state of charge of the battery and the charging time; S2: Obtain historical data on different types of EV charging at EV charging stations, and fit the probability distribution parameters of EV charging patterns to simulate and generate multiple random scenarios; Based on the simulated random scenarios, the mathematical model of the single EV's operating boundary is aggregated using the Minkowski summation method to obtain the operating boundary model of the EV cluster. S2 includes the following steps: S21: Obtain historical charging data of EV cluster users from EV charging stations to obtain daily charging pattern characteristics of EV cluster users, and classify EV cluster users according to the charging pattern characteristics. The charging pattern characteristics include start charging time, end charging time, initial charge level, and expected charge level; and the different types of EV users include at least private electric vehicles, ride-hailing electric vehicles, and electric buses; The probability distribution of EV charging pattern characteristics was obtained by fitting the data using Python. S22: Using the Monte Carlo simulation method, multiple random charging scenarios for EV charging stations at different times within a single day are generated based on the probability distribution of the EV charging pattern. S23: Based on the simulated random scenario, the mathematical model of the single EV operating boundary is aggregated using the Minkowski summation method to obtain the operating boundary model of the EV cluster. Furthermore, the operational boundary model of the EV cluster includes the time-series energy boundary model and the time-series power boundary model of the EV cluster; The expression for the time-series energy boundary model of the EV cluster is: In the formula: , These represent the upper and lower bounds of the energy for the EV cluster, respectively. Represented as the first EVs during the time period Whether the user is on the network is a state variable; This represents the maximum number of EVs within the EV cluster; The mathematical expression for the timing power boundary of the EV cluster is: In the formula: , These represent the upper and lower power bounds of the EV cluster, respectively. S3: Establish the flexibility constraints of the EV cluster based on the operational boundary model of the EV cluster; and based on the flexibility constraints of the EV cluster, construct a joint clearing model for the power system energy and ancillary services market in which the EV cluster participates, with the lowest power generation cost in the electricity market as the optimization objective. Furthermore, based on the joint clearing model, cost reduction items brought about by the flexibility provided by the EV cluster are added, and the joint clearing model is optimized to quantify the diverse flexibility resource potential of the EV cluster. S4: Based on the simulation and solution of various random scenarios, the optimized joint clearing model is solved, and the data results of the simulation and solution are used to realize the quantitative assessment of the multiple flexibility provided by EV clusters in participating in the electricity market.

2. The method for quantitatively evaluating the multi-dimensional flexibility of electric vehicle clusters according to claim 1, characterized in that, The mathematical model of the single EV operating boundary established in S1 includes the time-series energy boundary model and the time-series power boundary model of a single EV. And the expression for the time-series energy boundary model of a single EV is: In the formula: These represent the upper and lower bounds of the energy of a single EV, respectively. These represent the initial battery level of the EV when it enters the charging station and the expected battery level when it leaves; These represent the upper limit of the achievable capacity of the EV battery and the lower limit of the capacity that must be retained, respectively. These represent the maximum charging power and maximum discharging power of the EV, respectively. These represent the time periods when the EV enters and leaves the charging station, respectively. Represented as a single time interval step size; Indicates the current time period; The expression for the time-series power boundary model of a single EV is: In the formula: , These are represented as the upper and lower bounds of the energy EV, respectively. , They represent The upper and lower bounds of energy for a single EV during a given time period.

3. The method for quantitatively evaluating the multi-dimensional flexibility of electric vehicle clusters according to claim 2, characterized in that, S3 includes the following steps: S31: Considering the energy and power requirements of the power system's multi-dimensional flexibility at different time scales, establish the flexibility constraints of the EV cluster based on the operational boundary model of the EV cluster. The aforementioned multi-faceted flexibility refers to the power system's ability to economically utilize various flexible resources to cope with uncertainties in power sources, power grids, and loads in order to maintain a dynamic balance between power supply and demand. Furthermore, the response time requirement for the flexibility of the EV cluster is reflected by the upper limit constraint of the auxiliary services, and the expression for the upper limit constraint of the auxiliary services is as follows: In the formula: , These respectively represent the charging and discharging power adjustment speeds of the EV cluster providing upward and downward power to the EV cluster. Capacity of auxiliary services; , These respectively represent the charging and discharging power adjustment speeds of the EV cluster providing upward and downward power to the EV cluster. The upper limit of auxiliary services; The duration requirement for EV cluster flexibility is reflected by the capacity stacking constraint and power accumulation constraint of the ancillary services. The expression for the capacity stacking constraint is: In the formula: Indicates that the EV cluster is in The time period provides the first k Capacity of class-based auxiliary services; Indicates that the EV cluster is in The time period provides the first k Capacity of downstream auxiliary services; This indicates the lower bound of the power of the EV cluster; This represents the upper limit of the power of the EV cluster; , These represent the charging and discharging power of the EV cluster, respectively. Indicates the type of ancillary service; This represents the set of ancillary service types whose duration requirement is less than one time period; This represents the set of ancillary service types that require a duration of more than one time period. The power constraint takes into account the cumulative effect of providing ancillary services, and the power accumulation constraint is specifically as follows: when When the energy accumulation constraint is met, the expression is: In the formula: Indicates the step size for a single time period; Indicates the lower bound of the energy of the EV cluster; This represents the upper bound of the energy of the EV cluster; This is represented as the battery level of the EV cluster. Represented as The duration required for ancillary services, and ; , These represent the charging and discharging efficiencies of the EV cluster, respectively. when When the energy accumulation constraint is met, the expression is: In the formula: Represented as The duration required for ancillary services, and ; S32: Based on the flexibility constraints of the EV cluster, and with the goal of minimizing the generation cost of the power market, construct a joint clearing model for the power system energy and ancillary services market in which the EV cluster participates. Furthermore, the joint clearing model includes an objective function and constraints. The constraints include at least the operating constraints of thermal power units, system power balance constraints, and the flexibility constraints of EV clusters. It also adds cost reduction items generated by the flexibility resources provided by EV clusters, and optimizes the joint clearing model as a multi-dimensional flexibility indicator for quantifying EV clusters. The optimized expression for the joint clearing model is as follows: In the formula: Indicates thermal power unit exist The cost of generating electricity during a given period; This represents the startup status variable of the generator unit; This represents the shutdown state variable of the generator unit; Represents the total number of time periods , , Represented as the coefficients of a quadratic function of the power generation cost of thermal power units; , These represent the single start-up and shutdown costs for thermal power units, respectively. This represents the number of thermal power units; This indicates cost reductions resulting from flexible resources; Represented as Cost compensation coefficient for flexible resources; Represented as Flexible resources in Demand capacity for a given time period; Represented as EV cluster in Time slots available Total capacity of flexible resources.

4. The method for quantitatively evaluating the multi-dimensional flexibility of electric vehicle clusters according to claim 3, characterized in that, S4 includes the following steps: S41: In each random charging scenario, by calling the CPLEX optimization toolbox in Matlab, the optimized joint clearing model is optimized and simulated to solve based on the constraints, and the actual power / energy and available multi-dimensional flexibility capacity data of the EV cluster at different time periods within a single day are obtained in the corresponding random charging scenario. S42: Based on the multivariate flexibility capacity data provided by the EV cluster in the corresponding random charging scenario, the mean value of the probability distribution function of the flexibility capacity of the EV cluster at different times of the day is obtained by fitting the data using Python's fitter package. S43: The mean of the probability distribution function of the flexibility capacity of the EV cluster at different times within a single day is selected as the evaluation index of the flexibility potential, so as to realize the quantitative evaluation of the EV cluster's participation in the market at different times within a single day and the provision of diverse flexibility.

Images