Electric vehicle charging scheduling method and system oriented to vehicle network interaction strategy

Abstract

The invention belongs to the technical field of electric vehicle charging scheduling, and particularly discloses a vehicle network interaction strategy-oriented electric vehicle charging scheduling method and system, and the method comprises the following steps: determining EVA operation income data and user charging cost data; based on the EVA operation income data and the user charging cost data, establishing a target function by taking maximization of the EVA operation income and minimization of the user charging cost as targets; determining constraint conditions of the target function; and solving the target function based on the constraint condition to obtain an electric vehicle charging scheduling scheme. According to the method, the problem of large-scale electric vehicle access of the power grid can be solved, the V2G response willingness of the user can be accurately predicted, and the method can adapt to dynamic charging scheduling schemes of different scenes and requirements, so that the problems of economical efficiency and safe operation of power grid operation are solved.

Description

Electric vehicle charging scheduling method and system oriented towards vehicle-to-grid interaction strategy Technical Field

[0001] This invention belongs to the field of electric vehicle charging scheduling technology, specifically relating to electric vehicle charging scheduling methods and systems oriented towards vehicle-to-grid interaction strategies. Background Technology

[0002] With the rapid increase in the number of electric vehicles (EVs) and the sharp rise in charging load, the power distribution network is under unprecedented pressure. In particular, the large-scale, unregulated charging behavior of EVs has had a significant impact on the stable operation of the power system. The cumulative effect of charging loads concentrated during peak hours not only significantly exacerbates the peak-valley difference in the power grid, leading to capacity shortages and even overload risks in some areas, but also may drive up system operating costs and restrict the effective absorption of intermittent renewable energy.

[0003] To address this challenge, traditional charging dispatch strategies primarily rely on base electricity price signals, centralized control of load levels, or simple time-of-use pricing. While these strategies can mitigate load fluctuations to some extent, they have significant limitations. Firstly, traditional charging dispatch strategies are detached from actual user needs, affecting the effectiveness of Electric Vehicle Aggregators (EVAs) in optimizing dispatch and reducing user participation. Secondly, with the development of Vehicle-to-Grid (V2G) technology, EVs not only function as loads but also possess the potential to supply power to the grid as distributed mobile energy storage units. However, user participation in V2G exhibits significant individual differences and dynamic variability. Therefore, traditional charging dispatch strategies cannot accurately match the charging needs of individual users, impacting not only the final effectiveness of EVAs' optimized dispatch but also reducing user participation. Furthermore, existing dispatch strategies are inadequate in handling the individual differences and dynamic variability of V2G user participation. Summary of the Invention

[0004] The purpose of this invention is to provide an electric vehicle charging scheduling method and system for vehicle-to-grid (V2G) interaction strategies, which can solve the problem of power grids coping with large-scale electric vehicle access, accurately predict users' V2G response intentions, and adapt to dynamic charging scheduling schemes for different scenarios and needs, thereby solving the problems of power grid operation economy and safe operation.

[0005] To achieve the above objectives, the present invention employs the following technical solution: According to one aspect of the present invention, an electric vehicle charging scheduling method oriented towards vehicle-to-grid interaction strategy is provided, comprising the following steps: determining EVA operating revenue data and user charging cost data; the EVA operating revenue data includes the EVA's electricity purchase cost from the grid, electricity sales revenue, unit operation and maintenance cost, energy storage output operation and maintenance cost, and incentive revenue from participating in demand response; the user charging cost data includes the user charging electricity price, V2G response electricity price, EV charging demand, and EV participation in grid demand response; the EV participation in grid demand response is obtained based on a constructed user participation in V2G response willingness model; the user participation in V2G response willingness model is obtained based on a user charging behavior feature tag library; based on the EVA operating revenue data and user charging cost data, an objective function is constructed with the goal of maximizing EVA operating revenue and minimizing user charging cost; the constraints of the objective function are determined; the objective function is solved based on the constraints to obtain an electric vehicle charging scheduling scheme.

[0006] By adopting the above technical solution, user behavior can be accurately characterized through a user charging behavior feature tag library, which helps to accurately predict individual user charging needs and flexibility.

[0007] The establishment of the objective function can endogenously coordinate the core economic interests of the power grid side, the operator side, and the user side in the mathematical model, thereby unifying the potentially conflicting objectives of the parties under the same optimization framework, guiding the system to spontaneously find the optimal solution, and achieving a win-win situation for both economic and social benefits.

[0008] Based on the established user participation V2G response willingness model, the amount of electricity that EVs are willing to provide in grid demand response can be obtained. This allows for the precise quantification of the reverse power supply capability that each EV is willing to provide, thereby upgrading the traditional, extensive "adjustable load" estimation to a refined "willingness response" prediction. This effectively prevents user resistance caused by over-scheduling or ineffective scheduling and ensures the sustainability of V2G services.

[0009] According to one embodiment of the present invention, a method for constructing a user charging behavior feature tag library includes the following steps: acquiring the user's historical EV travel data, including charging start time, charging end time, charging power, charging start SOC (state of charge), charging end SOC, charging duration, and battery capacity; classifying the user's historical EV travel data to obtain feature variables of the user's charging behavior; using a fuzzy C-means clustering algorithm to cluster the feature variables of the user's charging behavior, dividing the user into different user clusters; defining feature tags for different user clusters based on the clustering results; the feature tags include charging time pattern, charging power pattern, charging location pattern, energy demand pattern, and price-sensitive pattern; and establishing a user charging behavior feature tag library based on the feature tags.

[0010] By adopting the above technical solution, the user's actual charging process and energy status can be quantitatively and multidimensionally restored, thus providing a solid data foundation for subsequent accurate analysis of the user's charging time habits, energy replenishment needs and willingness to participate in V2G response. This is the core input for realizing refined charging scheduling.

[0011] By integrating the above-mentioned multiple feature tags to establish a comprehensive user charging behavior feature tag library, a multi-dimensional and three-dimensional behavioral profile can be generated for each user or user group. This enables the vehicle-to-grid interactive scheduling system to go beyond a single dimension and perform multi-objective collaborative optimization (such as taking into account grid safety, economy and user satisfaction), and formulate a more intelligent, more humanized and more efficient global charging scheduling solution.

[0012] According to one embodiment of the present invention, a method for constructing a user participation V2G response intention model includes the following steps: selecting several feature tags based on a user charging behavior feature tag library; defining input fuzzy sets based on the selected feature tags, wherein each input fuzzy set corresponds one-to-one with the selected feature tags; defining fuzzy linguistic values ​​and membership functions within each input fuzzy set to describe the state of the input variables; determining a fuzzy inference rule base based on the fuzzy linguistic values ​​of each fuzzy set, used to obtain fuzzy inference conclusions based on the linguistic values ​​of each fuzzy set; aggregating and defuzzifying the fuzzy inference conclusions output by the activated rules to obtain an output fuzzy set, wherein the output fuzzy set is used to output the user's willingness to participate in V2G.

[0013] By adopting the above technical solution and utilizing fuzzy theory, clear but uncertain user characteristics (from a tag library) are transformed into fuzzy inputs. By simulating the reasoning rules of human thinking, a fuzzy conclusion on response intention is obtained, and finally, it is transformed into a clear and operable quantitative indicator.

[0014] According to one embodiment of the present invention, the input fuzzy set includes a first fuzzy subset, a second fuzzy subset, and a third fuzzy subset; the first fuzzy subset is used to characterize the user's sensitivity to electricity price subsidies, the second fuzzy subset is used to characterize the user's willingness to participate in V2G in response to parking time, and the third fuzzy subset is used to characterize the user's willingness to participate in V2G in response to SOC status; the output fuzzy set is used to describe the strength of the user's willingness to participate in V2G.

[0015] According to one embodiment of the present invention, the amount of electricity generated by EVs participating in grid demand response is obtained based on a pre-constructed user participation V2G response willingness model, including the following steps: characterizing user charging behavior according to a preset user charging behavior feature tag library and determining user feature tags; inputting user feature tags into the input fuzzy set of the user participation V2G response willingness model to obtain user participation V2G willingness variables; and predicting the amount of electricity generated by EVs participating in grid demand response based on user participation V2G willingness variables.

[0016] According to one embodiment of the present invention, the objective function is as follows: ; In the formula, T is the number of time periods within the vehicle-to-network interaction cycle, and t is the current time period. Generally, the vehicle-to-network interaction cycle is one day, and it is divided into 24 time periods, each lasting 1 hour. , , , , These are, respectively, the revenue from selling electricity to the grid by EVA during time period t, the incentive income from participating in demand response, the cost of purchasing electricity, the unit operation and maintenance cost, and the operation and maintenance cost of energy storage output; , These are the user charging electricity price and the V2G response electricity sales price for time period t, respectively. , These represent the EV charging demand during time period t and the electricity used for grid demand response, respectively.

[0017] According to one embodiment of the present invention, the constraints include power balance constraints, unit equipment operation constraints, and external power grid constraints; the power balance constraint formula is shown below:

[0018] In the formula, , These represent the electricity purchased from and sold to the grid during time period t, respectively. , and These represent the output of the photovoltaic unit, wind turbine unit, and gas turbine unit during time period t, respectively. , , and These represent the charging power of the energy storage unit, the discharging power of the energy storage unit, the charging power of the EV, and the discharging power of the EV, respectively, during time period t. The base electrical load for time period t; the operating constraint formulas for the unit equipment are shown below: In the formula, This refers to the output of the m-th unit; , These are the minimum and maximum output values ​​of the m-th generating unit, respectively; the external power grid constraint formulas are shown below: In the formula, , These represent the minimum and maximum output values ​​of the external power grid, respectively.

[0019] According to one aspect of the present invention, an electric vehicle scheduling device for a vehicle-to-grid (V2G) interaction strategy is provided, comprising: a data acquisition module for determining EVA operating revenue data and user charging cost data; the EVA operating revenue data includes the EVA's electricity purchase cost from the grid, electricity sales revenue, generator operation and maintenance cost, energy storage output operation and maintenance cost, and incentive revenue from participating in demand response; the user charging cost data includes the user charging electricity price, V2G response electricity price, EV charging demand, and EV participation in grid demand response; the EV participation in grid demand response is obtained based on a pre-constructed user participation in V2G response willingness model; the user participation in V2G response willingness model is obtained based on a user charging behavior feature tag library; an objective function construction module for constructing an objective function based on the EVA operating revenue data and user charging cost data, with the goal of maximizing EVA operating revenue and minimizing user charging cost; a constraint condition confirmation module for determining the constraints of the objective function; and a scheduling scheme confirmation module for solving the objective function based on the constraints to obtain an electric vehicle charging scheduling scheme.

[0020] According to one aspect of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements an electric vehicle charging scheduling method for a vehicle-to-grid interaction strategy according to any of the above embodiments.

[0021] According to one aspect of the present invention, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements an electric vehicle charging scheduling method for a vehicle-to-grid interaction strategy according to any of the above embodiments.

[0022] Compared with existing technologies, this invention has at least the following beneficial effects: 1. This invention provides an electric vehicle charging scheduling method oriented towards vehicle-to-grid (V2G) interaction strategies. Through data-driven profiling and intention reasoning, it proposes an interpretable charging profile and a joint representation method of V2G intention, which can accurately depict the charging behavior characteristics of individual users and quantify their V2G response intentions. This forms a collaborative optimization EVA-user bidirectional benefit scheduling method, which can adapt to dynamic charging scheduling schemes for different scenarios and needs, thereby solving the problems of grid operation economy and safe operation. Furthermore, it is effective in mining the value of flexible resources on the demand side, reducing the peak-valley difference in system operation and EV charging costs, providing considerable and controllable user support for new power systems, and improving the safety of power system operation.

[0023] 2. By clustering the characteristic variables of user charging behavior, this invention can effectively handle the complexity and uncertainty of user behavior data, and divide users into clusters with similar behavior patterns, rather than simply choosing one or the other. Therefore, when formulating V2G strategies, more targeted and flexible scheduling instructions can be adopted for different clusters, improving the feasibility of the strategy and user acceptance.

[0024] 3. This invention utilizes fuzzy theory to transform clear but uncertain user characteristics (from a tag library) into fuzzy inputs. By simulating the reasoning rules of human thinking, a fuzzy conclusion on response intention is derived, and finally, it is transformed into a clear and operable quantitative indicator.

[0025] The electric vehicle scheduling device, electronic device, and computer-readable storage medium for vehicle-to-grid interaction strategies provided by this invention also solve the problems raised in the background section. Attached Figure Description

[0026] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention. In the accompanying drawings: Figure 1 is a flowchart of the electric vehicle charging scheduling method for the vehicle-to-grid interaction strategy in Embodiment 1; Figure 2 is a flowchart of the method for constructing the user charging behavior feature tag library in Embodiment 1; Figure 3 is a histogram and its normal fitting curve of the EV charging start time in the EV charging behavior feature profile of a certain area in Embodiment 1; Figure 4 is a histogram and its log-normal fitting curve of the EV charging duration in the EV charging behavior feature profile of a certain area in Embodiment 1; Figure 5 is a scheduling block diagram of EV users participating in V2G response in a residential community in Embodiment 1; Figure 6 is a curve of electrical load, EV load, and total electrical load generated by the simulation example in Embodiment 1; Figure 7 is a curve of energy storage capacity and time-of-use electricity price generated by the simulation example in Embodiment 1; Figure 8 is the energy scheduling result of the optimized output of each unit generated by the simulation example in Embodiment 1; Figure 9 is a structural block diagram of the electric vehicle scheduling device for the vehicle-to-grid interaction strategy in Embodiment 2; Figure 10 is a schematic diagram of the electronic equipment in Embodiment 3.

[0027] Reference numerals in the attached figures: Electronic device 100; Memory 101; Processor 102; Computer program 103; Communication bus 104. Detailed Implementation

[0028] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0029] The following detailed description is exemplary and intended to provide further detailed explanation of the invention. Unless otherwise specified, all technical terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this invention is for describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention.

[0030] Example 1: An electric vehicle charging scheduling method for vehicle-to-grid (V2G) interaction strategies, as shown in Figure 1, includes the following steps: Determining EVA (Electric Vehicle Amplifier) ​​operating revenue data and user charging cost data; EVA operating revenue data includes the EVA's electricity purchase cost from the grid, electricity sales revenue, generator operation and maintenance costs, energy storage output operation and maintenance costs, and incentive revenue from participating in demand response; user charging cost data includes user charging electricity price, V2G response electricity price, EV charging demand, and EV participation in grid demand response; the EV participation in grid demand response is obtained based on a pre-constructed user participation in V2G response willingness model; the user participation in V2G response willingness model is obtained based on a user charging behavior feature tag library; based on the EVA operating revenue data and user charging cost data, an objective function is constructed with the goal of maximizing EVA operating revenue and minimizing user charging costs; the constraints of the objective function are determined; the objective function is solved based on the constraints to obtain an electric vehicle charging scheduling scheme.

[0031] The establishment of the objective function can endogenously coordinate the core economic interests of the power grid side, the operator side, and the user side in the mathematical model, thereby unifying the potentially conflicting objectives of the parties under the same optimization framework, guiding the system to spontaneously find the optimal solution, and achieving a win-win situation for both economic and social benefits.

[0032] Accurately characterizing user behavior through a user charging behavior feature tag library helps to accurately predict individual user charging needs and flexibility.

[0033] The user participation V2G response willingness model, based on a user charging behavior feature tag library, transforms the patterns hidden in massive historical behavioral data into insights into users' subjective psychological willingness. This establishes a logical chain from "objective behavioral data" to "subjective response willingness," enabling the dispatch system not only to know whether a user "can" participate but also to predict their "willingness" to participate, thus facilitating precise demand response. Based on the constructed user participation V2G response willingness model, the electricity volume of EVs participating in grid demand response can be accurately quantified, allowing for precise quantification of the reverse power supply capacity each EV is willing to provide. This elevates the traditional, coarse-grained estimation of "adjustable load" to a refined prediction of "willingness to respond," effectively preventing user resistance caused by over- or ineffective dispatch and ensuring the sustainability of V2G services.

[0034] Based on the established user participation V2G response willingness model, the amount of electricity that EVs are willing to provide in grid demand response can be obtained. In the optimization calculation, the reverse power supply capability that each EV is willing to provide can be accurately quantified. This improves the traditional, extensive "adjustable load" estimation to a refined "willingness response" prediction, effectively preventing user resistance caused by over-schedule or ineffective scheduling and ensuring the sustainability of V2G services.

[0035] The electric vehicle charging scheduling method for vehicle-to-grid interaction strategy includes the following steps: S1. Determine EVA operating revenue data and user charging cost data.

[0036] User charging cost data includes user charging electricity price, V2G response electricity price, EV charging demand, and EV participation in grid demand response.

[0037] EVA operating revenue data includes the cost of purchasing electricity from the grid, the revenue from selling electricity, the unit operation and maintenance costs, and the energy storage output operation and maintenance costs.

[0038] The amount of electricity generated by EVs participating in grid demand response is obtained based on a pre-constructed user participation V2G response willingness model; the user participation V2G response willingness model is obtained based on a user charging behavior feature tag library.

[0039] To assess the V2G participation potential of EVs, the first step is to establish a user charging behavior characteristic tag library to accurately predict the available capacity for user clusters to participate in V2G. The method for constructing the user charging behavior characteristic tag library includes the following steps: acquiring users' historical EV travel data, which includes user ID, charging start time, charging end time, charging power, starting SOC, ending SOC, charging duration, and battery capacity. This historical data can accurately depict users' charging time preferences and power demand characteristics.

[0040] The system categorizes users' historical EV travel data to obtain characteristic variables of user charging behavior. Fuzzy C-means clustering algorithm is used to cluster the characteristic variables of user charging behavior, dividing users into different user clusters. Based on the clustering results, feature labels are defined for different user clusters. Feature labels include charging time pattern, charging power pattern, charging location pattern, energy demand pattern, and price sensitivity pattern. A user charging behavior feature label library is established based on the feature labels.

[0041] The process of building a user charging behavior feature tag library based on historical data is divided into five steps: data collection, data preprocessing, feature calculation, user tag library, and profile construction. See Figure 2 for the specific process.

[0042] The process of building a user charging behavior feature tag library based on users' historical EV travel data mainly includes key steps such as data preprocessing, feature classification, feature calculation, establishing a user tag library, and constructing user profiles. The main implementation steps are as follows: 1) Data preprocessing, which mainly involves cleaning and normalizing missing values, outliers, aligning timestamps, and standardizing data formats to ensure data quality and consistency.

[0043] 2) Feature classification: Select key data dimensions that can reflect the characteristics of EV charging behavior, such as charging start time, charging end time, charging power, charging start SOC, charging end SOC, charging duration, and charging location.

[0044] Statistical analysis and calculations are performed on each user's multiple charging history data to generate the following characteristic variables: charging time characteristics: for example, the period when the user charges most frequently (such as the proportion of charging at night) and the average start time of charging.

[0045] Charging power characteristics: For example, the charging power level most frequently used by the user (slow charging / fast charging), and the average charging power.

[0046] Charging location characteristics: For example, the frequency of users charging at home, workplace, and public charging stations.

[0047] Energy demand characteristics: For example, the average energy replenished per charge and the average initial SOC (reflecting the user's "battery anxiety" level).

[0048] Price-sensitive characteristics: for example, the proportion of users who choose to charge during off-peak hours and their responsiveness to dynamic electricity prices.

[0049] 3) Feature calculation: The fuzzy C-means clustering algorithm is used to cluster the feature variables of user charging behavior, and users are divided into different user clusters.

[0050] 4) Establish a user tag library.

[0051] The cluster centers of each cluster (i.e., the average characteristics of users in that cluster) are analyzed. Based on the salient features of their feature vectors, descriptive feature labels are assigned to each user cluster, such as charging time patterns, charging power patterns, charging location patterns, energy demand patterns, and price sensitivity patterns. Specific user labels are defined for different user clusters based on the clustering results. In this way, user charging behavior is transformed into labels with practical scheduling guidance significance. Each user is associated with its cluster label to establish a queryable and updatable user charging behavior feature label library.

[0052] The ultimate goal is to provide accurate charging behavior profiles for EV users. Based on these profiles, we can obtain information such as user habits, charging start time, charging end time, charging power, and expected charge, thereby supporting differentiated optimization scheduling strategies, such as sending different electricity price incentives or V2G invitations to users with different tags.

[0053] Figure 3-4 shows a profile of EV charging behavior in a certain area. Figure 3 shows a histogram of the EV charging start time and its normal fit curve. Figure 4 shows a histogram of the EV charging duration and its log-normal fit curve.

[0054] By adopting the above technical solution, the user's actual charging process and energy status can be quantitatively and multidimensionally restored, thus providing a solid data foundation for subsequent accurate analysis of the user's charging time habits, energy replenishment needs and willingness to participate in V2G response. This is the core input for realizing refined charging scheduling.

[0055] By integrating the above-mentioned multiple feature tags to establish a comprehensive user charging behavior feature tag library, a multi-dimensional and three-dimensional behavioral profile can be generated for each user or user group. This enables the vehicle-to-grid interactive scheduling system to go beyond a single dimension and perform multi-objective collaborative optimization (such as taking into account grid safety, economy and user satisfaction), and formulate a more intelligent, more humanized and more efficient global charging scheduling solution.

[0056] By clustering the characteristic variables of user charging behavior, the complexity and uncertainty of user behavior data can be effectively handled. Users can be divided into clusters with similar behavioral patterns, rather than simply being either one or the other. Therefore, when formulating V2G strategies, more targeted and flexible scheduling instructions can be adopted for different clusters, improving the feasibility of the strategy and user acceptance.

[0057] Specifically, by using the fuzzy C-means clustering algorithm to perform feature calculation and clustering on users' historical data, the complexity and uncertainty of users' EV travel data can be effectively handled. Users are divided into clusters with similar behavioral patterns, rather than simple either / or approaches. Therefore, when formulating vehicle-to-everything (V2X) interaction strategies, more targeted and flexible scheduling instructions can be adopted for different clusters, improving the feasibility of the strategies and user acceptance.

[0058] By defining “charging time pattern” labels for different user clusters based on clustering results, the high-frequency charging time distribution of various types of users (such as nighttime home charging type, daytime office charging type, etc.) can be clearly identified. Thus, in vehicle-to-grid interaction, “nighttime charging type” users can be given priority to participate in valley filling, or “daytime charging type” users can avoid concentrated charging during peak electricity consumption, so as to achieve the goal of smoothing the load curve and peak shaving and valley filling.

[0059] By defining "charging power mode" labels for different user clusters based on clustering results, it is possible to distinguish users' dependence on and usage habits of charging facilities with different power levels, such as slow charging and fast charging. Thus, when the power grid needs to adjust the load, it can accurately schedule slow charging users who are not sensitive to power to charge in an orderly manner or operate at reduced power, while ensuring the experience of users with rigid demand for high power (such as long-distance travel refueling), thereby achieving optimal resource allocation.

[0060] By defining "energy demand pattern" labels for different user clusters based on clustering results, it is possible to quantify the energy replenishment expectations of users for a single charge and the daily operating range of battery SOC. Thus, in vehicle-to-grid interaction, the system can prioritize scheduling user vehicles with high energy demand elasticity and high SOC safety margin as distributed energy storage units for discharge (V2G) or adjust charging plans, maximizing the grid's regulation capacity while meeting users' core travel needs.

[0061] By defining “price-sensitive pattern” labels for different user clusters based on clustering results, it is possible to identify which user groups are more inclined to respond to electricity price signals to reduce charging costs. Thus, when formulating vehicle-grid interaction strategies, time-of-use pricing and real-time pricing economic incentive strategies can be designed and implemented for “high price-sensitive” user groups to effectively guide their charging behavior and transfer grid load at a lower cost.

[0062] In this way, when the power grid needs to adjust the load, precise scheduling can effectively guide the user group, maximize the power grid's regulation capacity, and achieve optimal resource allocation.

[0063] To accurately predict EV users' willingness to participate in V2G, a model can be built using a pre-constructed user charging behavior feature label library. This model is a fuzzy inference system (FIS) model built upon the pre-constructed user charging behavior feature label library. It is an effective method for handling uncertainty, utilizing expert knowledge, and providing highly interpretable results. The model includes the following steps: S1-11. Selecting several feature labels based on the user charging behavior feature label library, defining input fuzzy sets based on the selected feature labels, with each input fuzzy set corresponding one-to-one with the selected feature labels; defining fuzzy linguistic values ​​and membership functions within each input fuzzy set to describe the state of the input variables. In this embodiment, the input fuzzy sets include a first fuzzy subset, a second fuzzy subset, and a third fuzzy subset; the first fuzzy subset characterizes the user's sensitivity to electricity price subsidies, the second fuzzy subset characterizes the user's willingness to participate in V2G based on parking time, and the third fuzzy subset characterizes the user's willingness to participate in V2G based on SOC status.

[0064] Generally, S-type, Z-type, or Π-type membership functions are chosen. It should be noted that in actual processing, the parameters of the membership function can be set by experts based on experience. Furthermore, the parameters of the membership function need to be adjusted according to the mapping relationship between each input fuzzy subset and the output fuzzy set, or according to the semantics of the subsequently determined fuzzy inference rule base, to obtain a membership function that meets the requirements.

[0065] S1-22. Determine the fuzzy inference rule base based on the fuzzy linguistic values ​​of each fuzzy set, which is used to obtain fuzzy inference conclusions based on the linguistic values ​​of each fuzzy set.

[0066] S1-33. Aggregate and defuzzify the fuzzy inference conclusions output by the activated rules to obtain an output fuzzy set, which is used to output the variable of user willingness to participate in V2G.

[0067] Fuzzy inference generates multiple rules that may be partially activated, each outputting a fuzzy conclusion. These fuzzy conclusions are aggregated to form a comprehensive fuzzy output set. Through defuzzification (such as the centroid method), the aggregated fuzzy output is converted into a precise numerical value. , Let be the variable representing the willingness of the i-th user to participate in V2G, a response willingness score between 0 and 1, allowing the model's output to be quantified, understood, and applied. In this embodiment, the output fuzzy set is used to describe the strength of a user's willingness to participate in V2G. Through aggregation and defuzzification, the influence of all relevant rules can be integrated, ultimately outputting a precise response willingness value. This condenses the complex, multi-factor, and fuzzy user decision-making process into a key indicator that can be used for ranking, grouping, and policy triggering, enabling the V2G dispatch center to prioritize and mobilize "high-willing" user groups, significantly improving the success rate and efficiency of interactive responses.

[0068] Specifically, based on the EV user charging behavior feature tag library, several feature tags can be selected, such as "price-sensitive mode", "charging time mode", and "energy demand mode" as input variables for the model.

[0069] To characterize the impact of electricity price subsidies on user participation in V2G, three input fuzzy sets are defined in FIS. In the first fuzzy subset, three fuzzy subsets are defined, with corresponding fuzzy linguistic values ​​of "high," "medium," and "low," representing the sensitivity of EV users to electricity price subsidies; higher subsidies provide greater incentives and higher potential participation intentions, and vice versa. In the second fuzzy subset, five fuzzy subsets are defined, with corresponding fuzzy linguistic values ​​of "very long," "long," "medium," "short," and "very short," representing the influence of EV user parking time on their willingness to participate in V2G; longer parking times indicate a higher willingness to participate in V2G, and vice versa. In the third fuzzy subset, four fuzzy subsets are defined, with corresponding fuzzy linguistic values ​​of "very high," "high," "medium," and "low," measuring EV users' State of Charge (SOC) willingness to participate in V2G; higher SOC indicates a higher willingness to participate in V2G, and vice versa.

[0070] Five fuzzy subsets are defined in the output fuzzy set, with corresponding fuzzy linguistic values ​​of "positive", "high", "medium", "low" and "rejection", which are used to describe the strength of users' willingness to participate in V2G.

[0071] Fuzzy inference rules can be defined based on domain knowledge and business experience, enabling the model to simulate human decision-making processes. This allows the user participation V2G response willingness model to make complex judgments like human experts, accurately predicting the possible responses of different user groups to V2G incentive policies, and providing core decision support for developing personalized incentive schemes. A portion of the rule base is shown in Table 1.

[0072] Table 1. Rule Base (Partial)

[0073] The above complete fuzzy inference system requires 60 rules, but we can fill them with the full definition of the rule base. For combinations not listed, the fuzzy inference engine will activate multiple rules based on the membership degree of the input variables, and then perform aggregation and defuzzification.

[0074] Constructing a user participation V2G response intention model using a pre-built user charging behavior feature tag library is an effective way to transform complex and fuzzy user behavior characteristics into quantifiable V2G participation intentions. The core of this method lies in carefully designing a fuzzy rule base that reflects business logic, and simulating the expert decision-making process through steps such as fuzzification, rule evaluation, aggregation, and clarification, ultimately providing intuitive and powerful support for accurate V2G decision-making.

[0075] By adopting the above technical solution and utilizing fuzzy theory, clear but uncertain user characteristics (from a tag library) are transformed into fuzzy inputs. By simulating the reasoning rules of human thinking, a fuzzy conclusion on response intention is obtained, and finally, it is transformed into a clear and operable quantitative indicator.

[0076] S2. Based on EVA operating revenue data and user charging cost data, construct an objective function with the goal of maximizing EVA operating revenue and minimizing user charging cost.

[0077] Based on the aforementioned user charging behavior feature tag library and the analysis results of the user participation V2G response willingness model as a schedulable resource to participate in grid optimization scheduling, an optimization model is constructed with the goal of minimizing EV user charging costs and maximizing EVA operating benefits, while taking into account the physical constraints of EVs themselves and the users' willingness to participate.

[0078] S2-1. Based on the user charging behavior feature tag library, the EV charging behavior feature profile depicts the EV's travel time characteristics and power demand. By creating a feature profile of the user's charging behavior based on the user charging behavior feature tag library, and using historical data such as the user's EV-based charging start time, charging end time, and charging start SOC and charging end SOC, the user's expected SOC during charging can be obtained. Furthermore, based on the charging start SOC and expected SOC, the user's power demand can be determined.

[0079] The charging start time of an EV typically follows a normal distribution, and its probability density function is shown in equation (1) below: (1) Among them, This is the probability distribution function of the time when the EV connects to the power grid, which is also the probability distribution function of the start time of EV charging. The charging start time of the i-th EV Let be the mean of the probability function for the start of charging time. Let be the standard deviation of the probability function for the start time of charging.

[0080] Meanwhile, the charging departure time of the EV, that is, the time when the EV charging ends, usually conforms to a normal function distribution, and its related probability function density distribution is shown in the following equation (2): (2) Among them, Let be the time probability distribution function of the EV leaving the grid. The charging end time of the i-th EV Let be the mean of the probability function of charging completion over time. Let be the standard deviation of the probability function for the end of charging time.

[0081] The charging duration of an EV can be inferred from its charging start and end times, and an EV presence time matrix can be constructed. Based on the EV's charge level and charging power when it is connected to the grid, the expected charge level of the EV when it finishes charging can be inferred. Based on the expected charge level of the EV, the user's desired charge level can be adjusted, thereby obtaining the user's charge demand.

[0082] S2-2. The charging behavior characteristics of EV users can reflect the energy demand of EV users. After arriving at the charging location, EV users will charge the battery to a level higher than the expected level. The model can be built based on the EV user's arrival time, departure time, and expected SOC. Then, combined with the user's willingness to participate in V2G response model, the amount of electricity that EVs will participate in grid demand response can be predicted.

[0083] The amount of electricity generated by EVs participating in grid demand response is obtained based on a pre-constructed user participation V2G response willingness model, which includes the following steps: characterizing user charging behavior according to a preset user charging behavior feature label library and determining user feature labels; inputting user feature labels into the input fuzzy set of the user participation V2G response willingness model to obtain user participation V2G willingness variables; and predicting the amount of electricity generated by EVs participating in grid demand response based on user participation V2G willingness variables.

[0084] Based on the charging and discharging characteristics of EV batteries, the basic charging and discharging model of EV is constructed as shown in Equation (3), the battery capacity constraint of EV is shown in Equation (4), and the charging and discharging constraint variables of EV are shown in Equation (5).

[0085] (3) (4) (5) Among them, Let be the battery level of the i-th vehicle during time period t. Let be the initial battery level of the i-th EV. , These represent the charging and discharging power of the i-th EV during time period t. , These are the charging and discharging efficiencies of the EV, respectively. , 0 and 1 variables for EV charging and discharging.

[0086] (6) (7) (8) (9) (10) (11) Among them, The rated battery capacity of the i-th EV is , , These represent the charging time required for the i-th EV user to reach the desired battery level, the EV user's network access time, and the minimum time available for EVA scheduling to participate in V2G, respectively. , Let SOC and SOC be the expected SOC and the starting SOC of the i-th EV, respectively. The charging start time for the i-th EV. The charging end time for the i-th EV.

[0087] T represents the number of time periods within the vehicle-to-network interaction cycle, and t represents the current time period. Generally, the vehicle-to-network interaction cycle is one day, and it is divided into 24 time periods, each lasting 1 hour.

[0088] Let be the variable representing the willingness of the i-th user to participate in V2G, with a value range of [0,1]. This refers to the amount of electricity that the EV cluster can participate in grid demand response. The discharge threshold of the i-th EV It can be set according to user habits, generally speaking. It can be set to 35%-40%.

[0089] Let be the charging / discharging state variable of the i-th EV during the t-th time period. Its values ​​are -1, 0, and 1, which represent the EV discharging, being idle, and charging during the time period, respectively. , These represent the total charging power and total discharging power of the EV cluster participating in grid dispatch during time period t, respectively.

[0090] Specifically, by creating a feature profile of users' charging behavior based on a user charging behavior feature tag library, the expected SOC (State of Charge) of the i-th EV user during charging can be obtained. When the i-th EV is connected to the grid, the EV's starting state of charge (SOC) is obtained. And determine the rated battery capacity of the EV. Therefore, the charging time required for the i-th EV user to reach the desired battery level can be determined according to formula (6). .

[0091] Based on a user charging behavior feature tag library, a feature profile of user charging behavior is created, and a user participation V2G response intention model is constructed. Based on electricity price subsidies, EV SOC, and EV presence time matrix, the user's willingness to participate in V2G can be obtained, and the user's willingness to participate in V2G can be obtained. According to formula (9), the amount of electricity that the EV cluster can participate in grid demand response is obtained.

[0092] Based on a user charging behavior feature tag library, feature profiles can be created for user charging behavior, and the charging start time of the i-th EV user can also be determined. Charging end time The prediction is then performed, and the minimum duration for the i-th EV to be scheduled by EVA to participate in V2G is predicted according to formula (8). Based on the charging time required for the i-th EV user to reach the desired battery level. EV user network access duration Being able to infer the charging and discharging state of the i-th EV in time period t, that is, to determine The value of is then obtained according to formulas (10) and (11). , .

[0093] S2-3. Based on the above analysis, a model describing the charging behavior of EV users can be established. After further integrating factors related to user charging willingness, this model can distinguish between the dispatchable and undispatchable power of EV clusters. The implementation of this strategy can alleviate the power supply pressure on the grid caused by disorderly charging of EV clusters to a certain extent. Specifically, the operating revenue of EVA refers to the revenue from selling electricity to EV users plus the incentive income from participating in demand response, minus the cost of purchasing electricity from the grid, the unit operation and maintenance cost, and the energy storage output operation and maintenance cost, plus the EV user charging cost; while the charging cost of the EV cluster is defined as the total charging cost minus the incentive income obtained from participating in demand response. Therefore, maximizing the operating revenue of EVA is crucial. Minimum charging cost for EV users The objective functions are as follows: (12) and (13): (12) (13) Where T is the number of time periods within the vehicle-to-network interaction cycle, and t is the current time period; , , , , These represent, respectively, the revenue from selling electricity to the grid by EVA during time period t, the incentive income from participating in demand response, the cost of purchasing electricity, the unit operation and maintenance cost, and the operation and maintenance cost of energy storage output. , These are the charging electricity price for EV users and the electricity price for V2G response during time period t, respectively. , These represent the EV charging demand during time period t and the EV participation in grid demand response, respectively.

[0094] S3. Determine the constraints of the objective function.

[0095] The power balance constraint formula of the system is shown in equation (14) below: In formula (14), , These represent the electricity purchased from and sold to the grid during time period t, respectively. , and These represent the output of the photovoltaic unit, wind turbine unit, and gas turbine unit during time period t, respectively. , , and These represent the charging power of the energy storage unit, the discharging power of the energy storage unit, the charging power of the EV, and the discharging power of the EV, respectively, during time period t. The basic electrical load for time period t; the unit equipment operation and external power grid constraints are shown in (15) and (16) below: (15) (16) Among them, , These are the minimum and maximum output values ​​of the m-th unit, respectively. , These represent the minimum and maximum output values ​​of the external power grid, respectively.

[0096] S4. Solve the objective function based on the constraints to obtain the electric vehicle charging scheduling scheme.

[0097] In this way, under the premise of strictly meeting individual user constraints and system safety constraints, the globally optimal charging and discharging strategy is automatically generated, thereby transforming the complex multi-objective, high-dimensional, nonlinear scheduling problem into a computable mathematical model, providing dispatchers with clear, quantitative, and directly executable decision support, and improving the intelligence level of power grid operation control.

[0098] This embodiment uses actual charging data from EV users in a residential community as a case study. Addressing the challenges posed by the large-scale grid connection of electric vehicles, a charging physical process and V2G response willingness model for EV users are constructed based on a user charging behavior feature tag library and a user participation V2G response willingness model. Through simulation modeling and solution, a win-win situation is achieved by minimizing EV user charging costs and maximizing EVA (Electronic Vehicle Alliance) operating revenue. The EV user participation V2G response scheduling is shown in Figure 5.

[0099] The time-of-use pricing structure adopted is detailed in Table 2. This study aims to provide an effective strategy for managing large-scale EV access to the grid and coordinating user demand with grid operation by integrating user behavior modeling and cluster optimization solutions.

[0100] Table 2 Peak-Valley Electricity Prices

[0101] The core of this invention lies in constructing a three-in-one bidirectional optimization framework of "profile-willingness-scheduling," aiming to drive EV users with charging behavior profiles and willingness to participate in scheduling to efficiently participate in grid optimization solutions. This framework achieves deep integration of three key aspects: charging behavior profiles, scheduling willingness, and grid optimization solutions. It constructs an EV user aggregated scheduling model guided by dual-objective collaborative optimization on both the grid and EV sides, and solves for the optimal charging and discharging plan through efficient algorithms. This framework transforms behavioral data into profile modeling, combines willingness assessment to form a scheduling potential pool, performs optimization calculations based on real-time grid status and objectives, and feeds back the generated scheduling signals to users to trigger willingness responses, ultimately forming a closed-loop bidirectional interactive optimization mechanism.

[0102] Using the above methods, this experiment designed 150 EVs to participate in the EVA control. The system's scheduling energy can be provided based on the physical charging behavior characteristics of the EVs and the user's willingness to participate in V2G response. To simplify the solution, all EVs in this case study adopted uniform key parameter settings; specific values ​​are shown in Table 3. EV grid entry time follows a normal distribution with a mean of 18.6 and a variance of 5.6; EV grid exit time follows a normal distribution with a mean of 8 and a variance of 4; EV grid entry SOC follows a normal distribution with a mean of 0.6 and a variance of 0.2.

[0103] Table 3 EV Parameters

[0104] Figure 6 shows the electrical load, EV load, and total electrical load curves generated based on the EV charging behavior characteristic labels and the V2G response willingness model in the example simulation. Figure 7 shows the energy storage capacity and time-of-use electricity price curves. Figure 8 shows the energy dispatch results of the optimized output of each unit.

[0105] The electric vehicle scheduling method based on vehicle-to-grid (V2G) interaction strategy in this embodiment, under the scenario of large-scale centralized EV charging, creates a feature profile of EV user charging behavior based on a user charging behavior feature tag library, accurately characterizing the electricity consumption behavior features of EV users. Based on this feature profile, a user participation V2G response willingness model is constructed to fully assess the anxiety costs EV users face when considering grid dispatching. By fully considering user willingness, an incentive-based response scheduling strategy is proposed, improving EVA (Electronic Vehicle Allocation) operational efficiency and reducing EV user charging costs, while simultaneously achieving peak shaving and valley filling, promoting the economical and safe operation of the power grid.

[0106] This electric vehicle scheduling method, geared towards vehicle-to-grid (V2G) interaction strategies, comprehensively considers key influencing factors such as user sensitivity to electricity price subsidies, parking duration, and State of Charge (SOC). It achieves nonlinear mapping of multiple inputs through a fuzzy rule base and transforms fuzzy inference linguistic values ​​into quantifiable willingness indices through a defuzzification algorithm, thus achieving a leap from qualitative description to quantitative evaluation and providing a refined willingness assessment model for scheduling strategies. Deeply integrating the "user-centric" concept into V2G scheduling, it accurately portrays user behavior through multi-dimensional charging profiles and achieves refined quantification of V2G response willingness through fuzzy inference. Ultimately, it constructs a three-in-one bidirectional optimization framework of "profile-will-scheduling," maximizing the auxiliary value of V2G to the power grid while ensuring user travel needs. This overcomes the limitations of traditional scheduling methods that "emphasize the power grid over users" and "emphasize static over dynamic," providing a more feasible solution for the large-scale deployment of V2G technology.

[0107] The aforementioned method enables accurate resource prediction and proactive user participation in large-scale electric vehicle grid interaction. Its advantages lie not only in its multi-objective optimization capabilities at the technical level, but also in building a sustainable collaborative ecosystem of "user demand - grid demand." It dynamically optimizes the dual objectives of the grid and EV users. On the grid side, during peak load periods, it prioritizes the V2G discharge capacity of users with high willingness to participate, smoothing the load curve and alleviating grid pressure; during off-peak periods, it guides charging, improving the absorption rate of new energy sources. On the user side, differentiated incentive strategies are matched based on a willingness model to ensure user benefits and vehicle usage needs, reducing the decline in participation rates caused by user anxiety about charging.

[0108] Example 2: An electric vehicle scheduling device for vehicle-to-grid (V2G) interaction strategies, as shown in Figure 9, includes: a data acquisition module for determining EVA operating revenue data and user charging cost data; the EVA operating revenue data includes the EVA's electricity purchase cost from the grid, electricity sales revenue, generator operation and maintenance costs, energy storage output operation and maintenance costs, and incentive revenue from participating in demand response; the user charging cost data includes the user charging electricity price, V2G response electricity price, EV charging demand, and EV participation in grid demand response; the EV participation in grid demand response is obtained based on a pre-constructed user participation in V2G response willingness model; the user participation in V2G response willingness model is obtained based on a user charging behavior feature tag library; an objective function construction module for constructing an objective function based on the EVA operating revenue data and user charging cost data, with the goal of maximizing EVA operating revenue and minimizing user charging costs; a constraint condition confirmation module for determining the constraints of the objective function; and a scheduling scheme confirmation module for solving the objective function based on the constraints to obtain an electric vehicle charging scheduling scheme.

[0109] As shown in Figure 10, the present invention also provides an electronic device 100 for implementing an electric vehicle charging scheduling method for a vehicle-to-grid interaction strategy. The electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and capable of running on at least one processor 102, and at least one communication bus 104.

[0110] The memory 101 can be used to store the computer program 103. The processor 102 implements the steps of the method for analyzing and finding abnormal power line losses in Embodiment 1 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101.

[0111] The memory 101 may primarily include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created based on the use of the electronic device 100 (such as audio data), etc. In addition, the memory 101 may include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other non-volatile solid-state storage device.

[0112] At least one processor 102 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Processor 102 may be a microprocessor or any conventional processor. Processor 102 is the control center of electronic device 100, connecting various parts of electronic device 100 via various interfaces and lines.

[0113] The memory 101 in the electronic device 100 stores multiple instructions to implement an electric vehicle charging scheduling method oriented towards vehicle-to-grid interaction strategies. The processor 102 can execute multiple instructions to achieve the following: determining EVA operating revenue data and user charging cost data; the EVA operating revenue data includes the EVA's electricity purchase cost from the grid, electricity sales revenue, generator operation and maintenance costs, energy storage output operation and maintenance costs, and incentive revenue from participating in demand response; the user charging cost data includes the user charging electricity price, V2G response electricity price, EV charging demand, and EV participation in grid demand response; the EV participation in grid demand response is obtained based on a pre-constructed user participation in V2G response willingness model; the user participation in V2G response willingness model is obtained based on a user charging behavior feature tag library; based on the EVA operating revenue data and user charging cost data, an objective function is constructed with the goal of maximizing EVA operating revenue and minimizing user charging costs; the constraints of the objective function are determined; and the objective function is solved based on the constraints to obtain the electric vehicle charging scheduling scheme.

[0114] If the modules / units integrated in the electronic device 100 of Example 4 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, and read-only memory (ROM).

[0115] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0116] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.

[0117] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0118] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.

[0119] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0120] 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 it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. An electric vehicle charging scheduling method for vehicle-grid interaction strategies, characterized in that, Includes the following steps: The following steps are taken to determine EVA operating revenue data and user charging cost data: The EVA operating revenue data includes the cost of electricity purchased from the grid by the EVA, electricity sales revenue, unit operation and maintenance costs, energy storage output operation and maintenance costs, and incentive revenue from participating in demand response; the user charging cost data includes the user charging electricity price, V2G response electricity price, EV charging demand, and the amount of electricity EVs participate in grid demand response; the amount of electricity EVs participate in grid demand response is obtained based on a pre-constructed user participation in V2G response willingness model; this model is obtained based on a user charging behavior feature tag library; and based on the EVA operating revenue data and user charging cost data, an objective function is constructed to maximize EVA operating revenue and minimize user charging costs. Determine the constraints of the objective function; The objective function is solved based on the constraints to obtain the electric vehicle charging scheduling scheme.

2. The electric vehicle charging scheduling method based on vehicle-to-grid interaction strategy according to claim 1, characterized in that, The method for constructing the user charging behavior feature tag library includes the following steps: acquiring the user's historical EV travel data, which includes charging start time, charging end time, charging power, charging start SOC, charging end SOC, and battery capacity; classifying the user's historical EV travel data to obtain feature variables of the user's charging behavior; using a fuzzy C-means clustering algorithm to cluster the feature variables of the user's charging behavior, dividing the user into different user clusters; defining feature tags for different user clusters based on the clustering results; and establishing a user charging behavior feature tag library based on the feature tags.

3. The electric vehicle charging scheduling method based on vehicle-to-grid interaction strategy according to claim 1, characterized in that, The method for constructing the user participation V2G response intention model includes the following steps: selecting several feature labels based on a user charging behavior feature label library; defining input fuzzy sets based on the selected feature labels, wherein each input fuzzy set corresponds one-to-one with the selected feature labels; defining fuzzy linguistic values ​​and membership functions within each input fuzzy set to describe the state of the input variables; determining a fuzzy inference rule base based on the fuzzy linguistic values ​​of each fuzzy set, used to obtain fuzzy inference conclusions based on the linguistic values ​​of each fuzzy set; aggregating and defuzzifying the fuzzy inference conclusions output by the activated rules to obtain an output fuzzy set, which is used to output the user's willingness to participate in V2G.

4. The electric vehicle charging scheduling method based on vehicle-to-grid interaction strategy according to claim 3, characterized in that, The input fuzzy set includes a first fuzzy subset, a second fuzzy subset, and a third fuzzy subset; the first fuzzy subset is used to characterize the user's sensitivity to electricity price subsidies, the second fuzzy subset is used to characterize the user's willingness to participate in V2G in response to parking time, and the third fuzzy subset is used to characterize the user's willingness to participate in V2G in response to SOC status; the output fuzzy set is used to describe the strength of the user's willingness to participate in V2G.

5. The electric vehicle charging scheduling method based on vehicle-to-grid interaction strategy according to claim 3, characterized in that, The amount of electricity generated by EVs participating in grid demand response is obtained based on a pre-constructed user participation V2G response willingness model, including the following steps: characterizing user charging behavior according to a preset user charging behavior feature tag library and determining user feature tags; inputting user feature tags into the input fuzzy set of the user participation V2G response willingness model to obtain user participation V2G willingness variables; predicting the amount of electricity generated by EVs participating in grid demand response based on user participation V2G willingness variables.

6. The electric vehicle charging scheduling method based on vehicle-to-grid interaction strategy according to claim 1, characterized in that, The objective function is as follows: ； In the formula, T represents the number of time periods within the vehicle-to-network interaction cycle, and t represents the current time period. 、 、 、 、 These are, respectively, the revenue from selling electricity to the grid by EVA during time period t, the incentive income from participating in demand response, the cost of purchasing electricity, the unit operation and maintenance cost, and the operation and maintenance cost of energy storage output; 、 These are the user charging electricity price and the V2G response electricity sales price for time period t, respectively. 、 These represent the EV charging demand during time period t and the electricity used for grid demand response, respectively.

7. The electric vehicle charging scheduling method based on vehicle-to-grid interaction strategy according to claim 1, characterized in that, The constraints include power balance constraints, unit equipment operation constraints, and external power grid constraints; the power balance constraint formula is shown below: In the formula, 、 These represent the electricity purchased from and sold to the grid during time period t, respectively. 、 and These represent the output of the photovoltaic unit, wind turbine unit, and gas turbine unit during time period t, respectively. 、 、 and These represent the charging power of the energy storage unit, the discharging power of the energy storage unit, the charging power of the EV, and the discharging power of the EV, respectively, during time period t. The base electrical load for time period t; the operating constraint formulas for the unit equipment are shown below: In the formula, This refers to the output of the m-th unit; 、 These are the minimum and maximum output values ​​of the m-th generating unit, respectively; the external power grid constraint formulas are shown below: In the formula, 、 These represent the minimum and maximum output values ​​of the external power grid, respectively.

8. An electric vehicle dispatching device for vehicle-to-grid (V2G) interaction strategies, characterized in that, include: The data acquisition module is used to determine EVA operating revenue data and user charging cost data; The EVA operating revenue data includes the EVA's electricity purchase cost from the grid, electricity sales revenue, unit operation and maintenance costs, energy storage output operation and maintenance costs, and incentive revenue from participating in demand response; the user charging cost data includes the user charging electricity price, V2G response electricity price, EV charging demand, and EV participation in grid demand response; the EV participation in grid demand response is obtained based on a pre-constructed user participation in V2G response willingness model; the user participation in V2G response willingness model is obtained based on a user charging behavior feature tag library; the objective function construction module is used to construct an objective function based on the EVA operating revenue data and user charging cost data, with the goal of maximizing EVA operating revenue and minimizing user charging costs; The constraint verification module is used to determine the constraints of the objective function; The scheduling scheme confirmation module is used to solve the objective function based on constraints to obtain the electric vehicle charging scheduling scheme.

9. An electronic device, characterized in that, It includes a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement the electric vehicle charging scheduling method for vehicle-to-grid interaction strategy as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one instruction, which, when executed by a processor, implements the electric vehicle charging scheduling method for vehicle-to-grid interaction strategies as described in any one of claims 1 to 7.

Images