A method for coordinating and scheduling energy storage and V2G in a low-carbon urban park virtual power plant

Abstract

The present application relates to the technical field of virtual power plant optimization scheduling, and more particularly to a kind of energy storage and V2G collaborative scheduling method in urban low-carbon park virtual power plant, first, construct three kinds of electric vehicle user portraits of price-sensitive, time-sensitive, range anxiety, solve optimal charging path and output differentiated behavior constraints;Establish source-load-storage multi-resource coupling model, embed differentiated behavior constraints, design three kinds of energy storage-V2G collaborative operation mode and impose reverse collaborative constraints;Combined with carbon emission factor, establish park carbon emission calculation model, determine low-carbon objective function;Build three-objective optimization scheduling model of comprehensive operation cost, grid-connected load fluctuation rate and carbon emission, and solve, output optimal scheduling scheme considering economy, stability, low carbon and user satisfaction.Compared with prior art, the present application can improve park energy utilization efficiency, reduce carbon emissions, and protect user charging demand and grid stable operation.

Description

Technical Field

[0001] This invention relates to the field of virtual power plant optimization scheduling technology, and in particular to a method for coordinated scheduling of energy storage and V2G in a virtual power plant in a low-carbon urban park. Background Technology

[0002] In urban low-carbon parks, new energy resources such as distributed photovoltaics, energy storage, and V2G (Vehicle-to-Grid) electric vehicles are rapidly growing. Traditional dispatching models are gradually becoming inadequate to meet the operational needs of these new energy systems and are facing multiple bottlenecks. On the one hand, the behavioral characteristics of electric vehicle users vary significantly. Different users have significant differences in charging prices, time preferences, and range assurance. Disorderly charging behavior continues to exacerbate the peak-valley differences in park load, making it difficult to quantify the potential for flexible load adjustment. On the other hand, uncertainties on both the source and load sides are becoming increasingly prominent. The volatility of photovoltaic output and the randomness of electric vehicle load are superimposed, leading to a sharp increase in power fluctuations in the park and a significant increase in the risk to grid stability.

[0003] Furthermore, energy storage systems and V2G charging piles typically operate independently, lacking a unified coordination and control mechanism. This makes it difficult to effectively improve the absorption efficiency of renewable energy in the park, resulting in excess electricity that cannot be stored or deployed. Meanwhile, most existing dispatch schemes are designed around economic efficiency and grid stability, failing to incorporate carbon emission reduction as a core optimization objective. This is significantly out of sync with the increasingly stringent carbon reduction requirements of urban low-carbon parks, further limiting the efficient and low-carbon development of the park's energy system.

[0004] Existing technologies are mostly optimized for single resources or single objectives, failing to achieve refined modeling of user behavior, making it difficult to accurately predict and guide user behavior, unable to achieve deep synergy between energy storage and V2G, and unable to deeply integrate low-carbon goals with scheduling optimization. These shortcomings make existing technologies completely unsuitable for the efficient operation requirements of virtual power plants in urban low-carbon parks.

[0005] A search revealed Chinese invention patent application publication number CN112186809A, which discloses an optimal collaborative scheduling method for virtual power plants based on the V2G (Vehicle-to-Government) model of electric vehicles. The method establishes a charging station model considering the V2G model of electric vehicles; it establishes a mixed-integer linear programming model for optimal collaborative scheduling of virtual power plants based on the V2G model of electric vehicles, with the objective of maximizing the expected revenue of the day-ahead market and real-time market minus the operating cost of the virtual power plant; under the premise of satisfying the constraints of the virtual power plant and the electric vehicle, the mixed-integer linear programming model is solved, and based on the solution results, the energy sources within the virtual power plant are scheduled, the charging station is scheduled, and participation in market bidding is considered; the energy sources within the virtual power plant include electrical loads, photovoltaic power plants, and micro gas turbines. This existing patent application suffers from the problem of not considering user profiles and failing to realize the collaborative constraints between the energy storage system and V2G, resulting in poor collaborative scheduling efficiency.

[0006] How to achieve coordinated scheduling of energy storage and V2G in virtual power plants in urban low-carbon parks has become a technical problem that needs to be solved. Summary of the Invention

[0007] The purpose of this invention is to achieve refined characterization of electric vehicle user behavior and tap the potential of flexible load adjustment, construct a collaborative operation mechanism that efficiently complements energy storage and V2G, smooth out power fluctuations in the park's grid connection, improve the level of renewable energy consumption, and integrate carbon emission reduction indicators into the core dispatch system. Ultimately, it aims to achieve multi-objective collaborative optimization of the park's virtual power plant operation economy, grid stability, and decarbonization, and ensure the safe, efficient, and stable operation of the energy system in the city's low-carbon park.

[0008] The objective of this invention can be achieved through the following technical solutions: According to one aspect of the present invention, a method for coordinated scheduling of energy storage and V2G in a virtual power plant in an urban low-carbon park is provided. The method includes the following steps: based on the charging decision preferences of V2G users, constructing user profiles for electric vehicles including price-sensitive, time-sensitive, and range-anxiety users; using a dynamic Dijkstra algorithm to solve for the optimal charging paths for each type of user; constructing a comprehensive satisfaction index and outputting differentiated behavioral constraints; establishing a source-load-storage multi-resource coupling model; embedding the differentiated behavioral constraints into the source-load-storage multi-resource coupling model; and designing three types of energy storage including peak shaving and valley filling, fluctuation smoothing, and reserve capacity. The V2G collaborative operation mode defines the net load and applies reverse collaborative constraints to construct a collaborative constraint system based on user needs. Combining carbon emission factors and the aforementioned multi-resource coupling model, a carbon emission calculation model for the park is established, determining a low-carbon objective function centered on minimizing total carbon emissions. Integrating differentiated behavior constraints, the multi-resource coupling model, the collaborative operation mode, and the carbon emission calculation model, a three-objective optimization scheduling model is constructed, considering comprehensive operating costs, grid-connected load volatility, and carbon emissions. Solving this three-objective optimization scheduling model outputs the optimal collaborative scheduling scheme that balances economy, stability, low carbon emissions, and user satisfaction.

[0009] As a preferred technical solution, the process of solving the optimal charging path for various types of users includes: With minimizing charging costs as the core decision-making objective, we construct a user profile for price-sensitive individuals. With the shortest total travel time as the core decision-making objective, construct a time-sensitive user profile; With the core decision-making objective of ensuring the remaining driving range after charging, we construct a user profile for those with range anxiety. Differentiated path utility functions are constructed for different types of users, and the dynamic Dijkstra algorithm is used to solve for the optimal charging path for each type of user in the charging network.

[0010] As a preferred technical solution, the construction of the comprehensive satisfaction index specifically involves: using the extreme value normalization method to eliminate the dimensional differences between charging cost, time cost, and range anxiety, and constructing single-type satisfaction indices for each type of user; integrating the single-type satisfaction indices into a comprehensive satisfaction index through weighted summation; and constraining the user-side behavior boundaries that can participate in scheduling with the goal of maximizing comprehensive satisfaction.

[0011] As a preferred technical solution, the differentiated behavior constraints specifically include: determining the state of charge safety threshold, schedulable time period, and adjustable power range for various types of users based on various user profiles.

[0012] As a preferred technical solution, the establishment of the source-load-storage multi-resource coupling model is specifically as follows: Based on differentiated behavioral constraints and actual equipment parameters in the park, mathematical models of various resources within the virtual power plant are established, including photovoltaic power output model, micro gas turbine operation model, building load model, electric vehicle charging and discharging model, and energy storage system operation model. Based on the two-way time-of-use pricing mechanism, this paper integrates the cost items of grid interaction, micro-turbine, energy storage, and photovoltaic curtailment, and adds the revenue from selling electricity to building loads and ancillary services to construct a comprehensive operating cost model for virtual power plants. The grid-connected power fluctuation rate is introduced as an auxiliary optimization target to complete the multi-resource coupling model of source-load-storage.

[0013] As a preferred technical solution, the triggering logic and priority of the three energy storage-V2G collaborative operation modes are as follows: the backup capacity mode has the highest priority and is triggered first when the power shortage endangers the system safety; the fluctuation smoothing mode is the second highest and is triggered when the output fluctuation of renewable energy exceeds the set threshold; the peak shaving and valley filling mode has the lowest priority as the daily economic dispatch benchmark. All three modes must satisfy the reverse collaboration constraint and be arbitrated in order of priority.

[0014] As a preferred technical solution, the triggering mechanism for the standby capacity mode is as follows: If the triggering conditions for the peak shaving and valley filling mode are not met, and the triggering conditions for the fluctuation smoothing mode are also not met, and the virtual power plant experiences a power shortage, then the reserve capacity mode is triggered. The triggering conditions for the peak shaving and valley filling mode are as follows: when the net load of the virtual power plant is greater than or equal to the peak threshold, the peak shaving phase is triggered, and energy storage and V2G are given priority to discharge; when the net load of the virtual power plant is less than the valley threshold, the valley filling phase is triggered, and energy storage and V2G are given priority to charge. The triggering condition for the fluctuation smoothing mode is as follows: if the triggering condition for the peak shaving and valley filling mode is not met, and the variance of the grid-connected power sliding window is detected to be greater than the set value or the rate of change of grid-connected power at adjacent times exceeds the ramp rate limit, then the fluctuation smoothing mode is triggered.

[0015] As a preferred technical solution, the specific steps for defining the net load and applying reverse collaborative constraints are as follows: Define the net power of the energy storage system and the net power of the electric vehicle respectively, and apply the reverse cooperative constraint so that the signs of their net power are opposite at any scheduling time; The reverse coordination constraint includes a reverse coordination depth coefficient and a reverse coordination depth constraint. The reverse coordination depth coefficient is the ratio of the smaller of the absolute value of the net power of the energy storage system and the absolute value of the net power of the electric vehicle at the same time to the larger of the two values, and the denominator is avoided to be zero by a preset minimum positive number. The larger the value, the higher the degree of reverse matching between the power of the energy storage and the electric vehicle. The reverse coordination depth constraint is that the reverse coordination depth coefficient at any time is not lower than a preset reverse coordination depth threshold.

[0016] As a preferred technical solution, the comprehensive objective function of the three-objective optimization scheduling model is a weighted fusion of the economic objective function value, the grid-connected load volatility objective function value, and the low-carbon objective function value. The economic objective function value is based on minimizing the full life cycle operating cost of the park, including grid interaction cost, micro gas turbine operating cost, energy storage system depreciation cost, photovoltaic curtailment penalty cost, and reverse collaboration penalty cost, while deducting building load electricity sales revenue and ancillary service revenue.

[0017] As a preferred technical solution, the objective function value of the grid-connected load volatility is the variance between the grid-connected power and the average grid-connected power within the scheduling period.

[0018] Compared with the prior art, the present invention has the following beneficial effects: 1) This invention creates differentiated user profiles based on the charging decision preferences of V2G users, embeds differentiated behavioral constraints into a multi-resource coupling model, and designs a collaborative operation mode that includes peak shaving and valley filling, fluctuation smoothing, and reserve capacity. It is deeply integrated with low-carbon target optimization, realizing a closed-loop control of the entire process from user-side behavioral constraints to park-level three-objective optimization. It effectively solves the problem of disconnect between user experience and system benefits in traditional scheduling, and provides a systematic technical solution for balancing park economy, stability, low carbon emissions, and user satisfaction.

[0019] 2) This invention constructs three types of differentiated user profiles and solves the optimal charging path, thereby achieving refined modeling of the charging decision preferences of different users. This not only ensures the charging cost for price-sensitive users and the travel efficiency for time-sensitive users, but also meets the range safety needs of range-anxious users. It clarifies the specific content of differentiated behavioral constraints, ensuring the basic charging needs of users and providing a clear user-side adjustable boundary for energy storage-V2G collaborative operation. This effectively improves the engineering feasibility and user acceptance of subsequent collaborative scheduling schemes and avoids the drawbacks of neglecting user needs in traditional scheduling.

[0020] 3) This invention constructs a source-load-storage multi-resource coupling model. On the one hand, by modeling photovoltaic, micro gas turbine, load, electric vehicle and energy storage resources separately, it realizes the accurate quantification of the operating characteristics of various energy equipment in the park. On the other hand, by integrating cost items and revenue items and introducing grid-connected power fluctuation rate target, the multi-resource model is coupled into a unified whole that can be used for optimized scheduling, providing reliable underlying model support for subsequent collaborative operation mode and three-objective optimization, and effectively improving the economy and operational stability of the scheduling scheme.

[0021] 4) This invention constructs three energy storage-V2G collaborative operation modes: peak shaving and valley filling, fluctuation smoothing, and reserve capacity mode. By clearly defining triggering conditions, it achieves accurate mode identification and switching under different operating scenarios. By setting mode priorities and reverse collaborative constraints, it solves the problem of multi-mode triggering conflicts, ensuring the orderly distribution of power and the effectiveness of collaboration. It not only achieves full-coverage operation control of the park in scenarios such as daily economic dispatch, power fluctuation suppression, and safety reserve guarantee, but also avoids frequent mode switching and command conflicts, significantly improving the stability, security, and engineering practicality of system operation.

[0022] 5) The three-objective optimization scheduling model of this invention integrates the economic objective function, the grid-connected load volatility objective function, and the low-carbon objective function in a weighted manner. The economic objective function considers the cost of reverse coordination penalty, while the grid-connected load volatility objective function considers the variance between the grid-connected power and the average grid-connected power within the scheduling period to suppress random fluctuations in grid-connected power. The integrated objective function achieves unified quantification and weighted optimization of the three dimensions of park operation economy, grid interaction stability, and low-carbon objectives, solving the problem of difficulty in coordinating and balancing multiple objectives in traditional scheduling. It provides complete objective function support for subsequent solutions to obtain the optimal scheduling scheme that takes into account the benefits of all parties. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating the collaborative scheduling method in this invention; Figure 2 These are the time-series curves of total grid load under four scenarios in this invention; Figure 3 These are the timing curves of the power exchange between the microgrid and the power grid under four scenarios in this invention; Figure 4 This is the timing sequence of energy storage and V2G charging and discharging power in this invention; Figure 5 This invention provides a comparison of user satisfaction for the same type of user across four different scenarios. Detailed Implementation

[0024] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0025] The purpose of this invention is to achieve refined characterization of electric vehicle user behavior and tap the potential of flexible load adjustment, construct a collaborative operation mechanism that efficiently complements energy storage and V2G, smooth out power fluctuations in the park's grid connection, improve the level of renewable energy consumption, and integrate carbon emission reduction indicators into the core dispatch system. Ultimately, it aims to achieve multi-objective collaborative optimization of the park's virtual power plant operation economy, grid stability, and decarbonization, and ensure the safe, efficient, and stable operation of the energy system in the city's low-carbon park.

[0026] The innovative aspects of this invention include: 1. Construct three types of electric vehicle user profiles and output differentiated behavioral constraints: Based on the charging decision preferences of V2G users, construct three types of user profiles: price-sensitive, time-sensitive, and range-anxiety-prone. Use the dynamic Dijkstra algorithm to solve the optimal charging path for each type of user, construct a comprehensive satisfaction index, and output differentiated behavioral constraints such as the state of charge safety threshold, schedulable time period, and adjustable power range to achieve refined modeling of user-side behavior and quantification of adjustable potential.

[0027] 2. Design three energy storage-V2G collaborative operation modes and apply reverse collaborative constraints: Establish a source-load-storage multi-resource coupling model, design three collaborative operation modes: peak shaving and valley filling, fluctuation smoothing, and reserve capacity, and clarify the triggering logic and priority of each mode; define the net power of energy storage and the net power of electric vehicles, apply reverse collaborative constraints, and introduce a reverse collaborative depth coefficient and depth constraint to avoid ineffective power flow and improve the effectiveness of collaboration.

[0028] 3. Determine the low-carbon objective function based on the carbon emission calculation model of the park considering carbon emission factors: Combine the carbon emission factors of power grid purchase, natural gas carbon emission factors and emission reduction of photovoltaic substitution, establish a carbon emission calculation model of the park, determine the low-carbon objective function with the minimization of total carbon emissions as the core, and integrate the low-carbon objective with economic objectives and grid-connected load volatility objectives to achieve deep integration of low-carbonization and dispatch optimization.

[0029] 4. Construct a three-objective optimization scheduling model for comprehensive operating cost, grid-connected load volatility, and carbon emissions: Integrate differentiated behavior constraints, multi-resource coupling model, collaborative operation mode, and carbon emission calculation model to construct a three-objective optimization scheduling model. The model integrates the economic efficiency, stability, and low carbon emissions of the comprehensive objective function, and uses an improved particle swarm optimization algorithm to solve the problem, outputting the optimal collaborative scheduling scheme that takes into account economic efficiency, stability, low carbon emissions, and user satisfaction.

[0030] Example 1 This embodiment relates to a method for coordinated scheduling of energy storage and V2G in a virtual power plant in a low-carbon urban park, such as... Figure 1 The method of the present invention includes the following steps: S1. Collect multi-source operational data of the virtual power plant in the urban low-carbon park, including electric vehicle charging data, distributed photovoltaic time-series output data, energy storage system operating parameters, V2G charging pile parameters, building rigid load time-series data, two-way time-of-use electricity pricing and ancillary service pricing data, as well as grid carbon emission factors and natural gas carbon emission factors. The collected data undergoes preprocessing, including cleaning, outlier removal, and extreme value normalization, to form a standardized time-series dataset, providing a data foundation for subsequent modeling.

[0031] S2. Based on the preprocessed electric vehicle travel and charging data from S1, users are segmented according to their core preferences in charging choices, thereby completing the accurate construction of electric vehicle user profiles. This includes the following sub-steps: S2.1. Users are divided into three typical categories: price-sensitive, time-sensitive, and mileage-anxious. S2.2. For the three types of users, a differentiated path utility function is constructed based on the core decision-making objective, and the dynamic Dijkstra algorithm is used to determine the optimal charging path and charging node: (1) Price-sensitive users: These users prioritize minimizing charging costs. Referring to the current time-of-use pricing policy, the charging price at each time period at each road network node is used as the main basis for path weighting. Users choose to charge during periods with lower electricity prices based on their grid connection time to achieve optimal charging costs.

[0032] , In the formula, Indicates the first i electric vehicles t The cost of charging at any time is the core optimization goal for price-sensitive users; express t The real-time electricity price at any given moment, referencing the current time-of-use pricing policy, is the main basis for path weighting, and users will choose to charge during periods with lower electricity prices; Indicates the first i electric vehicles t The charging power at any given time; Indicates the first i The charging time of an electric vehicle represents the length of time required to complete this charging process.

[0033] (2) Time-sensitive users: These users make decisions based on minimizing travel time. Since road conditions change dynamically over time, the time required to reach the destination varies, and their charging solutions prioritize the path with the shortest total travel time.

[0034] , In the formula, Indicates the first ielectric vehicles t The time cost at any given moment is a decisive factor for time-sensitive users in choosing their routes, and minimizing this metric is crucial. express t From road network nodes i To the node j The travel time for a given road segment varies over time as road conditions change dynamically.

[0035] (3) Range-anxious users: These users have significant concerns about battery charge depletion, tend to choose the shortest route, and require the state of charge (SOC) to remain above the safe threshold. When SOC is low, they prioritize finding the nearest charging station to recharge. This invention uses a power function to describe the evolution of range anxiety within a driving cycle, thus more realistically reflecting the psychological characteristic that anxiety accelerates as the remaining range decreases. By minimizing the total range anxiety generated during charging, the model effectively improves the overall satisfaction of this type of user.

[0036] , In the formula, Indicates the first i electric vehicles t The range anxiety score at any given moment quantifies the user's level of concern about insufficient battery life, and its value ranges from [value range missing]. ; This represents the user's range anxiety level during the driving period, with a maximum value of 1. l Indicates the current mileage traveled; Indicates the user's comfort mileage threshold; This indicates the remaining range when the battery is fully charged.

[0037] , In the formula, The range efficiency coefficient is determined by the energy consumption per unit distance of the electric vehicle. It is numerically equal to 1 / energy consumption per unit distance and reflects the driving range that can be converted from each unit of battery capacity. C This indicates the rated battery capacity of the EV.

[0038] S2.3. The extreme value normalization method is used to eliminate the dimensional differences in charging cost, time cost, and range anxiety. First, a single-type satisfaction model for each type of user is constructed, and then a comprehensive satisfaction model is obtained through weighted summation. (1) Single-type satisfaction model: The core cost and anxiety indicators of each type of user are normalized.

[0039] , In the formula, , , These are individual satisfaction metrics for price-sensitive, time-sensitive, and mileage-anxious users, respectively. , , and , , These represent the maximum and minimum values ​​of charging cost, time cost, and range anxiety, respectively, obtained through historical data statistics.

[0040] (2) User comprehensive satisfaction model: By summing the weighted coefficients, the single-type satisfaction indexes of the three types of users are integrated into a unified comprehensive satisfaction index.

[0041] , In the formula, For overall satisfaction; a , b , c For the weighting coefficients, satisfying a + b + c = 1 These correspond to the satisfaction weights of price-sensitive, time-sensitive, and mileage-anxious users, respectively.

[0042] S2.4 Based on the electric vehicle user profile and comprehensive satisfaction model, the core preferences of different types of users are transformed into objective constraints that can be directly invoked by the scheduling model. Specifically, this includes outputting differentiated behavioral constraints such as the state of charge safety threshold, schedulable time period, and adjustable power range for each type of user, and clarifying the boundaries that users can participate in scheduling.

[0043] S3. Based on the multi-source operational data collected in S1, and combined with the electric vehicle user profiles and differentiated behavioral constraints output in S2, a virtual power plant source-load-storage multi-resource coupling model is constructed, including distributed photovoltaics, energy storage systems, V2G electric vehicles, building rigid loads, and micro gas turbines. Simultaneously, differentiated behavioral constraints are embedded into the entire scheduling process of the source-load-storage multi-resource coupling model, clarifying the overall operational constraints and providing boundary support for subsequent optimized scheduling. Specifically, this includes the following sub-steps: S3.1. Based on the actual equipment parameters and differentiated behavioral constraints of the park, mathematical models of each resource within the virtual power plant are established to achieve a full-dimensional characterization of source, load, and storage: (1) Distributed photovoltaic power output model: Considering the temporal effects of light intensity and temperature on photovoltaic power output, a power output prediction model is constructed as follows: , In the formula, express t The actual output power of the photovoltaic system at any given time; Indicates the rated power under standard test conditions; express tThe actual light intensity at any given moment; Indicates the light intensity under standard test conditions; Indicates the power temperature coefficient of a photovoltaic module; express t The operating temperature of the photovoltaic modules at any given time; This indicates the component temperature under standard test conditions.

[0044] (2) Micro gas turbine operation model. This model focuses on gas consumption rate and power generation efficiency, and constructs an output power model and an efficiency correction model as follows: , In the formula, express t Output electrical power at any given moment; express t Power generation efficiency at any given time; express t Gas consumption rate at any given time; Indicates the lower heating value of the gas; Indicates rated efficiency; Indicates the efficiency decline rate coefficient; This represents the power ratio corresponding to the point of highest efficiency. This indicates the rated power of the gas turbine.

[0045] (3) Building load model, which decomposes the building load into basic rigid load and random fluctuation component: , In the formula, express t Total load power at any given time; Indicates the base load; This represents the random fluctuation component, and its predicted value is used in optimized scheduling by fitting historical data. Indicates the first k Rated power of rigid equipment; Indicates the first k Rigidity in t The start / stop status at any given time (value is 0 or 1); N This indicates the number of types of rigid equipment within a building.

[0046] (4) V2G electric vehicle charging and discharging model: , In the formula, express t The state of charge of the battery at any given time; , express t The charging and discharging power at any given moment; , Indicates charging and discharging efficiency; This indicates the total battery capacity of the electric vehicle; Indicates the time step.

[0047] (5) Energy storage system operation model, which is based on the dynamic change law of battery state of charge (SOC) and constructs the energy balance equation for the charging and discharging process: , In the formula, express t The remaining battery power at any given time; , Indicates charging and discharging power; , Indicates charging and discharging efficiency; E EV (t) represents the total capacity of the electric vehicle battery; Indicates the time step of the scheduling.

[0048] S3.2 Considering the bidirectional load characteristics of photovoltaic and V2G systems, a corresponding bidirectional time-of-use pricing agreement needs to be established. This agreement stipulates that, within the same time period, the price at which the virtual power plant system purchases electricity from the grid is slightly higher than the price at which it sells electricity to the grid. Based on this bidirectional time-of-use pricing mechanism, a comprehensive operating cost model for the virtual power plant is constructed by integrating costs related to grid interaction, micro-turbines, energy storage, and photovoltaic curtailment, and adding revenue items such as building electricity sales and ancillary services, thus quantifying the system's full-cycle economic indicators.

[0049] (1) Integrated operating cost model: The core economic objective is to minimize the total operating cost of the virtual power plant, with positive cost and negative revenue.

[0050] , In the formula, This represents the total cost of joint operation of the virtual power plant; a negative value indicates net revenue. This indicates the net expenditure related to interaction with the power grid; This indicates the operating cost of a micro gas turbine; This indicates the depreciation cost of the energy storage system; This indicates the cost of penalties for curtailing solar power. This indicates revenue from selling electricity to building loads; This refers to revenue from ancillary services, including revenue generated from electric vehicle (EV) discharge and energy storage discharge.

[0051] (2) Power grid interaction cost model: Based on two-way time-of-use pricing, the purchase price of electricity is higher than the sales price of electricity. Peak shaving and valley filling and economic benefits are achieved through off-peak charging and peak discharge.

[0052] , In the formula,H Indicates the total number of time periods within the scheduling period; , They represent t The power purchased from and sold to the grid by the time-of-use microgrid meets the requirements. ; , They represent t Electricity purchase price and electricity sales price for the specified time period; This represents the net expenditure on interaction with the power grid.

[0053] (3) Operating cost model of micro gas turbine: The operating cost of micro gas turbine depends on the natural gas consumption.

[0054] , In the formula, This indicates the operating cost of a micro gas turbine; This represents the output electrical power of the micro gas turbine at time t; This indicates the power generation efficiency of a micro gas turbine; Indicates the price of natural gas; Indicates the scheduling time step.

[0055] (4) Depreciation cost model of energy storage system: The depreciation cost of energy storage system is related to the total charge and discharge volume and is calculated based on the unit depreciation cost. The expression is as follows: , In the formula, This indicates the depreciation cost of the energy storage system; , Let represent the charging power and discharging power of the energy storage system at time t, respectively; This represents the unit depreciation cost of energy storage.

[0056] (5) Photovoltaic curtailment penalty cost model: The revenue loss caused by photovoltaic curtailment is calculated based on the amount of curtailment and the on-grid benchmark electricity price.

[0057] , In the formula, This indicates the cost of penalties for curtailing solar power. Indicates the amount of solar power curtailed; This indicates the local benchmark on-grid electricity price for photovoltaic power.

[0058] (6) Building load electricity sales revenue model: the revenue obtained from selling electricity to building loads , In the formula, This indicates revenue from selling electricity to building loads; express tElectricity sold to building loads during specific time periods; This indicates the unit price of electricity sold to the building's load.

[0059] (7) Ancillary service revenue model: derived from the discharge behavior of electric vehicles and energy storage, calculated based on the unit price of ancillary services such as participation in grid peak-valley regulation.

[0060] , In the formula, Indicates revenue from ancillary services; This indicates the unit price of peak-valley regulation ancillary services; , Let represent the discharge power of the electric vehicle and the discharge power of the energy storage system at time t, respectively.

[0061] S3.3 To avoid drastic power fluctuations caused by simply pursuing economic optimization, grid-connected power fluctuation rate is introduced as an auxiliary optimization objective. A comprehensive objective function is constructed through weighted normalization to balance economy and stability.

[0062] (1) Grid-connected power fluctuation model: If only economic efficiency is the objective, since the electricity price is constant during peak, valley, and flat periods, the system may experience random fluctuations in charging and discharging power within the same electricity price period, while the economic target value remains the same, thus failing to effectively constrain load fluctuations. To suppress load fluctuations, a fluctuation penalty term is introduced, and the fluctuation rate is incorporated into the scheduling model as an auxiliary optimization objective through a weighted method. Specifically, the variance between grid-connected power and average power is used to quantify the degree of fluctuation; the greater the fluctuation, the higher the value of the penalty term.

[0063] , In the formula, This represents the target value for grid-connected power fluctuation rate; Represents the variance of grid-connected power; This represents the exchange power between the microgrid and the power grid during time period t. This represents the average grid-connected power during the scheduling period.

[0064] (2) Normalized comprehensive objective function: In order to eliminate the difference in dimensions between comprehensive cost and grid-connected power fluctuation rate, the weighted coefficient method is used to transform the dual objective optimization into a single objective problem.

[0065] , In the formula, This represents the normalized comprehensive objective function value. , These represent the weighting coefficients for overall cost and load volatility, respectively, satisfying... ; , and , These represent the maximum and minimum reference values ​​for comprehensive cost and the maximum and minimum reference values ​​for grid-connected power fluctuation, respectively. and These represent the objective function values ​​for overall cost and grid-connected power volatility, respectively. By adjusting the weighting coefficients, different optimization effects can be achieved, balancing economic efficiency and stability.

[0066] S3.4 To ensure the safe and stable operation of the virtual power plant, rigid constraints are set across the entire domain from three dimensions: power balance, equipment power, and state of charge. All decision variables must meet the constraint requirements.

[0067] (1) Power balance constraint: All power generation, discharge, grid interaction and power consumption loads in the system at any time t Maintain power balance.

[0068] , (2) Equipment power constraints: The output or input power of each device is limited by its rated capacity and operating status.

[0069] , In the formula, This indicates the maximum output power of the photovoltaic array; , These represent the minimum stable operating power and rated power of the micro gas turbine, respectively.

[0070] V2G electric vehicle charging and discharging power constraints: The charging and discharging power of V2G electric vehicles must be limited by their maximum charging and discharging power, and charging and discharging states are mutually exclusive within the same time period. The constraint expression is as follows: , In the formula, , This indicates the maximum charging and discharging power of the V2G electric vehicle; , A binary variable representing the charging and discharging status of a V2G electric vehicle, with a value of 0 or 1.

[0071] Energy storage system charging and discharging power constraints: The charging and discharging power of the energy storage system must be limited by its maximum charging and discharging power, and charging and discharging states are mutually exclusive within the same time period. The constraint expression is as follows: , In the formula, , This indicates the maximum charging and discharging power of the energy storage system; , A binary variable representing the charging and discharging state of an energy storage system, with a value of 0 or 1.

[0072] Power grid interaction constraints: The power purchased and sold between the virtual power plant and the power grid is limited by their interaction capacity. The constraint expression is as follows: , In the formula, , This indicates the maximum power purchased from or sold to the power grid.

[0073] (3) EV and energy storage SOC constraints: The state of charge of electric vehicles and energy storage systems must be maintained within the safe operating range, and terminal constraints may be added.

[0074] V2G electric vehicle SOC constraints: To ensure user travel needs and battery safety, the state of charge (SOC) of electric vehicles must meet upper and lower limits as well as terminal SOC constraints, expressed as follows: , In the formula, This represents the state of charge of the V2G electric vehicle in time period t. , Indicates the SOC limit for electric vehicles; This represents the remaining battery power of the electric vehicle in time period t. Indicates the rated capacity of the electric vehicle battery; T This indicates the time at the end of the scheduling cycle.

[0075] Energy storage system SOC constraints: To ensure the safe operation and cycle life of the energy storage system, the state of charge of the energy storage system must meet the upper and lower limit constraints and the terminal state of charge constraints, expressed as follows: , In the formula, This represents the state of charge of the energy storage system during time period t. , Indicates the SOC limit for energy storage; This represents the remaining power of the energy storage system during time period t. Indicates the rated capacity of the energy storage battery; , These represent the state of charge of the energy storage system at the beginning and end of the scheduling cycle, respectively.

[0076] S4. Based on the differentiated behavioral constraints of S2 and the multi-resource coupling model constructed in S3, with the core constraints of meeting user travel needs, SOC safety boundaries, and charging preferences, three complementary and coordinated operation modes of energy storage-V2G are designed, including peak shaving and valley filling, fluctuation smoothing, and reserve capacity. This achieves time-series reverse coordinated adjustment of charging and discharging power considering differences in user behavior. Specifically, it includes the following sub-steps: S4.1. Combining the differentiated behavioral constraints output by S2, and based on the virtual power plant source-load-storage multi-resource coupling model constructed by S3, three complementary and collaborative operation modes of energy storage system and V2G electric vehicle are designed.

[0077] (1) Peak shaving and valley filling mode: In response to the long-cycle fluctuation characteristics of the daily load curve, differentiated scheduling is implemented based on differentiated behavioral constraints: During peak load periods, the energy storage system and V2G electric vehicles discharge synchronously to jointly support the load gap. During scheduling, priority is given to ensuring the off-grid range needs of time-sensitive users, while reducing the amount of electricity purchased from the high-carbon grid and reducing the carbon emissions of the system; during off-peak periods and periods of high photovoltaic output, the energy storage system charges to absorb surplus photovoltaic power, and V2G electric vehicles charge opportunistically to reduce user charging costs. This mode aims to reduce the peak-valley difference of the system, improve the photovoltaic absorption rate, and protect the economic rights and interests of users, thereby achieving low-carbon and economically optimized system operation.

[0078] (2) Fluctuation Smoothing Mode: Targeting short-term peaks / dips in charging load and photovoltaic output, this mode focuses on rapid stabilization and implements instantaneous compensation within the user's dispatchable window. When an instantaneous power surge occurs, the energy storage system acts as the main rapid response entity to provide power support or absorption; the V2G electric vehicle dynamically adjusts according to the objective constraints of the S2 output (dispatchable time period and state-of-charge safety threshold). This mode aims to smooth grid-connected power fluctuations and avoid impacting the power grid. While ensuring grid stability, it effectively improves the user's operating experience through flexible adjustments that do not affect user travel.

[0079] (3) Backup Capacity Mode: Based on the backup capacity requirements reserved for user travel, and with the real-time power balance of the system as the core, a hierarchical backup response mechanism is constructed: the energy storage system serves as the primary backup capacity, quickly supplementing the power deficit or absorbing surplus power when there are deviations in photovoltaic output prediction, insufficient V2G response, or limited ramping of micro gas turbines; the V2G electric vehicle serves as an auxiliary backup, participating in the adjustment according to the scheduling priority generated by the user profile when the energy storage backup capacity is insufficient, prioritizing the use of vehicles that do not affect travel. The scheduling priority corresponding to the user profile is determined by a combination of parameters such as user type, schedulable time period, state of charge safety threshold, and adjustable power range output by S2. This mode aims to ensure the safe and stable operation of the system, taking into account both user travel needs and system power supply reliability under extreme conditions, providing a safety net for the low-carbon and economical operation of the system.

[0080] All three collaborative operation modes described above rely on the core mechanism of reverse collaboration between the energy storage system and the V2G electric vehicle. Specifically, when the energy storage system discharges, the V2G electric vehicle charges; when the energy storage system charges, the V2G electric vehicle discharges. To achieve this reverse collaboration between the charging and discharging power of the V2G electric vehicle and the energy storage system, the net power of the V2G electric vehicle is defined. With net energy storage power .

[0081] , In the formula, Indicates the number of electric vehicles participating in the scheduling; , Let represent the total discharge power and total charging power of all V2G electric vehicles at time t, respectively. , Indicates the first i electric vehicles at all times t The charging and discharging power (kW) is positive, and the charging power is negative. , Let represent the discharge power and charging power of the energy storage system at time t, respectively, with discharge power being positive and charging power being negative.

[0082] Under all operating modes, the reverse coordination regulation requires that the net power direction of the energy storage and the V2G electric vehicle be opposite, that is, to meet the reverse coordination constraint and ensure that there is no invalid power flow of charging and discharging in the same direction between the energy storage and the V2G electric vehicle at any time.

[0083] , To further quantify the degree of collaboration, ensure the engineering practicality of the collaboration mechanism, and avoid ineffective operation states where peak-shaving and valley-filling modes or fluctuation-smoothing modes only satisfy reverse constraints but lack sufficient collaboration depth, a reverse collaboration depth coefficient is introduced. And reverse collaborative depth constraints.

[0084] , In the formula, This represents a pre-defined, extremely small positive number, used to avoid mathematical anomalies where the denominator is zero; , representing the reverse cooperative depth coefficient at time t. The larger the value, the higher the degree of reverse power matching between energy storage and V2G electric vehicles, and the closer the reverse collaboration. This represents the threshold for reverse coordination depth, which is adaptively adjusted based on the park's load fluctuation level and grid operation requirements. It is increased when the park's net load fluctuates drastically or the grid-connected power fluctuation rate is large. Forced energy storage and V2G can achieve deeper reverse synergy to quickly smooth out fluctuations and reduce grid impact; when the system is operating smoothly, the load can be appropriately reduced. To enhance the flexibility of some flexible resources and prioritize economic efficiency, the park's operational status is categorized into three levels: high volatility, medium volatility, and low volatility. A high volatility state is defined as when the net load volatility or grid-connected power volatility is greater than or equal to a first threshold; a medium volatility state is defined as when the net load volatility or grid-connected power volatility is less than the first threshold but greater than or equal to a second threshold; and a low volatility state is defined as when both the net load volatility and grid-connected power volatility are less than the second threshold. Net load volatility is the ratio of the standard deviation of the current period's net load sequence to the mean net load, and grid-connected power volatility is the ratio of the standard deviation of the current period's grid-connected power sequence to the mean grid-connected power.

[0085] When the park's operation is in a highly volatile state, the reverse collaboration depth threshold will be adjusted to a preset high value to force energy storage and V2G electric vehicles to achieve high-depth reverse collaboration, so as to quickly smooth out load and power fluctuations and reduce grid impact. When in a state of medium fluctuation, the reverse collaboration depth threshold is adjusted to the preset median value to ensure the collaboration effect while taking into account the flexibility of park operation. When in a low-fluctuation state, the reverse collaboration depth threshold is adjusted to a preset low value to release the flexible adjustment space of energy storage and V2G electric vehicles, and prioritize the economic efficiency of park operation. Threshold smooth transition: A first-order low-pass filtering algorithm is used to smooth the adjusted reverse collaboration depth threshold, avoiding frequent switching of control commands caused by threshold step changes, and ensuring stable execution of the collaborative operation mode.

[0086] S4.2 To reflect the dynamic switching process of the system under different operating states, a conditional logic chain is used to describe the operating criteria and corresponding power allocation rules for peak shaving and valley filling mode, fluctuation smoothing mode, and reserve capacity mode, and the system net load is defined. .

[0087] , (1) Peak shaving and valley filling mode , In the formula, when When it enters the peak shaving and valley filling mode, the peak shaving phase begins; when When it enters the valley filling stage of peak shaving and valley filling mode; when At the same time, maintain the state of the previous stage to avoid frequent mode switching. , This indicates the upper limit of the total discharge and charging power of energy storage and V2G; , This indicates the peak and valley thresholds for the peak shaving and valley filling mode.

[0088] (2) Fluctuation Smoothing Pattern If none of the above conditions for peak shaving and valley filling are met (i.e.) Furthermore, abnormal fluctuations in grid-connected power were detected, specifically the variance of the grid-connected power sliding window. If the rate of change of grid-connected power exceeds the ramp rate limit at adjacent times, the fluctuation smoothing mode is triggered. In this mode, energy storage prioritizes responding to high-frequency fluctuations, while V2G responds to mid-frequency fluctuations, and the power is adjusted in a coordinated manner to meet the following requirements: , In the formula, Indicates the target grid-connected power. , These represent the regulation power of energy storage and V2G electric vehicles, respectively.

[0089] (3) Standby capacity mode If the above-mentioned peak shaving and valley filling conditions and fluctuation smoothing conditions are not met, and the system experiences a power deficit, i.e., real-time backup demand... If so, the standby capacity mode will be triggered.

[0090] , Energy storage systems prioritize providing backup capacity: , If the energy storage backup capacity is insufficient, i.e. Then, V2G electric vehicles will be used to provide backup capacity: , In the formula, This represents the backup capacity that the energy storage system can provide at time t. Indicates the minimum allowable capacity of the energy storage system; This indicates the maximum discharge power of the energy storage system. This represents the remaining energy of the energy storage system at time t; This indicates the lower limit of the safe state of charge of the energy storage system; Indicates the rated capacity of the energy storage battery; This represents the total reserve capacity available from all V2G electric vehicles at time t. This indicates the number of V2G electric vehicles participating in the scheduling; This represents the maximum discharge power of the k-th electric vehicle; This represents the remaining battery power of the k-th electric vehicle at time t; Indicates the first k The minimum permissible battery capacity for an electric vehicle; Indicates the first k The dispatchable status of an electric vehicle is determined by the dispatchable time period in the user profile and the SOC safety lower limit. A value of 1 indicates that the vehicle is in a dispatchable state, while a value of 0 indicates that it is not dispatchable.

[0091] S4.3 To avoid command conflicts between different modes, the system schedules each mode according to its priority: Peak shaving and valley filling mode serves as the daily economic scheduling benchmark and has the lowest priority; Fluctuation smoothing mode has a medium priority, temporarily taking over the adjustment right when excessive fluctuations are detected, and returning to the original mode after completion; Reserve capacity mode has the highest priority, triggered only when power shortages endanger system safety, and the V2G resources called at this time must meet the SOC safety lower limit and scheduling priority set by the user profile, and return step by step after execution. At each scheduling moment, the system first checks whether the triggering conditions of reserve capacity mode are met. If not, it further checks whether the triggering conditions of fluctuation smoothing mode are met. If neither is met, peak shaving and valley filling mode is run.

[0092] S5. Based on the multi-resource coupling model of source, load and storage in S3 and the carbon emission factor data collected in S1, establish a carbon emission calculation model for the park and construct a low-carbon objective function with the minimization of total carbon emissions as the core.

[0093] S5.1 Based on the energy consumption characteristics of the park, define a calculation model for the park's carbon emissions during the scheduling cycle.

[0094] , In the formula, Indicates the scheduling period H Total carbon emissions; Indicates the scheduling period H Total indirect carbon emissions corresponding to electricity purchases from the domestic power grid; Indicates the scheduling period H The total direct carbon emissions generated by the consumption of natural gas by internal micro gas turbines; Indicates the scheduling period H The total carbon emission reduction resulting from replacing fossil fuels with photovoltaic power generation.

[0095] , In the formula, This represents the average carbon emission factor of the power grid, which is determined based on the local power grid structure. Indicates the carbon emission factor of natural gas; This represents the electrical power purchased by the system from the grid during time period t; This represents the natural gas consumption of the micro gas turbine during time period t; Indicates time t The actual photovoltaic power absorbed is the photovoltaic output minus the curtailed portion: This carbon emission calculation model fully considers the coupling relationship between photovoltaic consumption, grid power purchase and gas turbine output, providing a precise quantitative basis for optimizing low-carbon targets; Indicates the scheduling time step.

[0096] S5.2 Based on the carbon emission calculation model, the goal function for low carbon emissions is determined to be minimizing total carbon emissions.

[0097] , in The objective function is low carbon emissions.

[0098] To eliminate dimensional differences and facilitate subsequent weighted integration with the economic objectives of S3.2 and the stability objectives of S3.3, the low-carbon objective function is subjected to weighted normalization: , In the formula, This represents the normalized low-carbon objective function value. , These represent the minimum and maximum reference values ​​for total carbon emissions within the scheduling cycle, respectively.

[0099] Based on the differentiated user behavior constraints of S2, the multi-resource coupling model and operation constraints of S3, the energy storage V2G collaborative operation mechanism of S4, and combined with the carbon emission calculation model and low-carbon objective function of S5, a three-objective optimization scheduling model that comprehensively considers economy, stability and low carbon is constructed.

[0100] S6.1 The economic objective is to minimize the full life cycle operating cost of the park, including grid interaction costs, micro gas turbine operating costs, energy storage system depreciation costs, photovoltaic curtailment penalty costs, and reverse collaboration penalty costs, minus building load electricity sales revenue and ancillary service revenue.

[0101] , In the formula, The cost of reverse collaboration penalty is expressed as: , In the formula, This represents the collaborative penalty weight (element), used to balance collaborative effects and economic efficiency; Reverse collaboration depth coefficient.

[0102] S6.2 To eliminate the dimensional differences among the three objectives of economy, stability, and low carbon and achieve weighted fusion, a linear extreme value normalization method is adopted. A three-objective optimization scheduling model that comprehensively considers economy, stability, and low carbon is constructed, and its comprehensive objective function expression is: , In the formula, , , Represents the weighting coefficients, satisfying ; , , These are the normalized economic objective function value, the grid-connected load volatility objective function value, and the low-carbon objective function value, respectively. , They represent the scheduling period respectively. H Minimum and maximum reference values ​​for internal comprehensive costs.

[0103] S6.3 To ensure the engineering feasibility of the optimization results, set constraints based on the constraints in S2-S4.

[0104] S7. Based on the actual park operation data collected and preprocessed in S1, the differentiated behavioral constraints, multi-resource coupling model, energy storage V2G complementary and collaborative operation mode, carbon emission calculation model and three-objective optimization scheduling model constructed in S2 to S6 are integrated into a complete optimization problem. Four sets of progressive comparative simulation scenarios are set up, and the improved particle swarm algorithm is used to solve the problem. The optimal collaborative scheduling scheme of the virtual power plant that takes into account economy, stability and low carbon emissions is output.

[0105] An improved particle swarm optimization algorithm is used to solve the problem. The improvements include the integration of various strategies such as chaotic initialization and elite reverse learning, adaptive inertial weights, non-dominated sorting and Pareto archive maintenance, hybrid constraint processing, hierarchical coding, and adaptive mutation. This enables the efficient and stable solution of the complex multi-objective and multi-constraint optimization problem of energy storage and V2G collaborative scheduling in the virtual power plant of urban low-carbon park, and outputs the final optimal collaborative scheduling scheme.

[0106] Example 2 This embodiment also relates to a method for coordinated scheduling of energy storage and V2G in a virtual power plant in a low-carbon urban park, including: Based on the three-objective optimization scheduling model, four progressive comparative simulation scenarios were set up: Scenario (1): The baseline scenario serves as a basic operational reference. It excludes electric vehicles, solar-energy storage charging stations, and distributed photovoltaic systems, and only includes residential and office building electrical loads and micro gas turbines. All loads operate according to fixed time-series curves without any flexible adjustment capabilities. This scenario is used to quantify the impact of flexible resource access on load characteristics, operational economics, and carbon emission levels, providing an evaluation benchmark for subsequent optimization strategies.

[0107] Scenario (2): Economic optimization scenario, with the single optimization objective of minimizing the overall operating cost, introduces basic collaboration between energy storage system and V2G charging pile, sets V2G penetration rate at 50%, does not distinguish user behavior type, and uses unified scheduling constraints for all electric vehicles to verify the adjustment effect of basic economic optimization.

[0108] Scenario (3): Low-carbon user behavior adaptation scenario. Based on scenario (2), user behavior modeling and carbon emission targets are introduced. The user composition ratio is set as follows: 40% price-sensitive, 30% time-sensitive, and 30% range-anxiety-sensitive. Differentiated behavioral constraints are set for the three types of users, namely, the lower limit of SOC for range-anxiety-sensitive users is increased to 35%, the off-grid time for time-sensitive users is strictly fixed, and the charging cost weight is added to the objective function for price-sensitive users. This is used to quantify the independent impact of user behavior differences and low-carbon targets on scheduling results and to verify the effectiveness of refined user modeling and carbon emission reduction constraints.

[0109] Scenario (4): Energy storage-V2G deep collaboration scenario. Based on the user behavior differentiation constraints in scenario (3), an energy storage-V2G reverse collaboration mechanism is introduced, including a reverse collaboration depth coefficient and hierarchical constraints, and a grid-connected power stability target is added to form a three-objective optimization model of economy, low carbon, and stability. Three complementary collaborative operation modes, namely peak shaving and valley filling, fluctuation smoothing, and reserve capacity, are adopted to verify the comprehensive improvement benefits of the deep collaboration mechanism on system economy, stability, low carbon and user satisfaction under user profile constraints.

[0110] Based on the above scenario, we integrate differentiated behavioral constraints, multi-resource coupling models, three complementary and collaborative operation modes, and three objective optimization functions, and use an improved particle swarm optimization algorithm to solve the optimization problem.

[0111] The simulation results of each scenario are compared and analyzed from five dimensions: economy, stability, collaborative effectiveness, low carbon emissions, and user experience. The focus is on verifying the comprehensive benefits of user profile modeling and deep collaborative mechanisms.

[0112] Figure 2 The graph shows a comparison of the total load of the park under four scenarios, with the horizontal axis representing the scheduling period (in 15 minutes) and the vertical axis representing the load power (in kW). Among them, the load curve of scenario 1 fluctuates the most, with a significant peak-to-valley difference; scenario 2 achieves initial peak shaving and valley filling through economic optimization; scenario 3, under the constraints of user behavior and low-carbon goals, further smooths the load curve; scenario 4, through deep collaboration between energy storage and V2G, achieves the optimal load curve smoothing effect, with a significant reduction in the peak-to-valley difference, demonstrating the optimization effect of multi-mode collaboration on load characteristics.

[0113] Figure 3The graphs show the power exchange curves between the microgrid and the grid in four scenarios. The horizontal axis represents the dispatch period (in 15 minutes), and the vertical axis represents the power exchange between the microgrid and the grid (in kW). In scenario 1, the grid-connected power fluctuates drastically, with obvious peaks and troughs. In scenario 2, the grid-connected power fluctuations are reduced through basic collaborative optimization. In scenario 3, under the constraints of user behavior and low-carbon goals, the peak-to-trough difference in grid-connected power is further reduced. In scenario 4, through the coordinated adjustment of the fluctuation smoothing mode and the reserve capacity mode, the grid-connected power curve is the smoothest, effectively reducing the impact on the grid and demonstrating the effect of multi-objective optimization on improving the stability of system operation.

[0114] Figure 4 The graph shows the power coordination curves for energy storage and V2G charging / discharging in Scenario 4. The horizontal axis represents the scheduling period, and the vertical axis represents the device charging / discharging power. The dashed line represents the V2G charging / discharging power curve, and the solid line represents the energy storage charging / discharging power curve. The curves show that, under the reverse coordination mechanism, the charging / discharging power directions of energy storage and V2G are always opposite: when energy storage is discharging, V2G is charging; when energy storage is charging, V2G is discharging. This achieves reverse power matching, effectively avoiding ineffective power flow during unidirectional charging and discharging, and verifying the effectiveness of the reverse coordination mechanism.

[0115] Figure 5 A bar chart comparing user satisfaction with different types of electric vehicles under four scenarios is presented. The horizontal axis represents the scenario type, and the vertical axis represents user satisfaction (values ​​range from 0 to 1, with higher values ​​indicating better user experience). The results show that: Scenario 1 (baseline scenario) has no V2G scheduling, and the satisfaction levels of all three types of users are at a low level. Scenario 2 (Economic Optimization Scenario) adopts a unified scheduling constraint, without distinguishing user behavior, resulting in limited improvement in satisfaction for users with mileage anxiety and time sensitivity. In scenario 3 (low-carbon user behavior adaptation scenario), after introducing differentiated behavioral constraints, the satisfaction of all three types of users improved significantly. Among them, the satisfaction of range-anxious users improved the most because the lower limit of SOC was guaranteed. Scenario 4 (Deep Collaboration between Energy Storage and V2G) achieved optimal scheduling based on differentiated constraints, with the satisfaction levels of all three types of users reaching the highest level, verifying the comprehensive improvement effect of user profile modeling and multi-mode collaboration mechanism on user experience.

[0116] Among them, the peak shaving and valley filling mode is used to verify the effects of long-term load regulation, photovoltaic absorption and low-carbon operation; the fluctuation smoothing mode is used to verify the effect of smoothing short-term fluctuations in instantaneous grid-connected power; and the reserve capacity mode is used to verify the system's power deficit compensation and safe and stable operation capabilities, thereby verifying the effectiveness of the optimization model and the three cooperative operation modes proposed in this invention.

[0117] Example 3 The electronic device of this invention includes a central processing unit (CPU), which can perform various appropriate actions and processes according to computer program instructions stored in read-only memory (ROM) or loaded from a storage unit into random access memory (RAM). The RAM may also store various programs and data required for device operation. The CPU, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0118] Multiple components in the device are connected to the I / O interface, including: input units such as keyboards and mice; output units such as various types of displays and speakers; storage units such as disks and optical discs; and communication units such as network interface cards (NICs), modems, and wireless transceivers. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0119] The processing unit executes the various methods and processes described above. For example, in some embodiments, the methods may be implemented as computer software programs tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the device via ROM and / or a communication unit. When the computer program is loaded into RAM and executed by the CPU, the energy storage and V2G collaborative scheduling method in a virtual power plant in a low-carbon urban park, as proposed in Embodiment 1, can be implemented. Alternatively, in other embodiments, the CPU may be configured to execute the methods by any other suitable means (e.g., by means of firmware).

[0120] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0121] The program code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0122] In the context of this invention, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0123] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method for coordinated scheduling of energy storage and V2G in a virtual power plant in a low-carbon urban park, characterized in that, The method includes the following steps: Based on the charging decision preferences of V2G users, a user profile of electric vehicles is constructed, including price-sensitive, time-sensitive, and range-anxiety-oriented users. The dynamic Dijkstra algorithm is used to solve for the optimal charging path for each user type, constructing a comprehensive satisfaction index and outputting differentiated behavioral constraints. A source-load-storage multi-resource coupling model is established, embedding the differentiated behavioral constraints into the model. Three energy storage-V2G collaborative operation modes are designed, including peak shaving and valley filling, fluctuation smoothing, and reserve capacity. Net load is defined and reverse collaborative constraints are applied, constructing a collaborative constraint system based on user demand. Combining carbon emission factors and the multi-resource coupling model, a carbon emission calculation model for the industrial park is established, determining a low-carbon objective function centered on minimizing total carbon emissions. Combining differentiated behavioral constraints, the multi-resource coupling model, collaborative operation modes, and the carbon emission calculation model, a three-objective optimization scheduling model is constructed, considering comprehensive operating cost, grid-connected load volatility, and carbon emissions. Solving the three-objective optimization scheduling model outputs the optimal collaborative scheduling scheme that balances economy, stability, low carbon emissions, and user satisfaction.

2. The method for coordinated scheduling of energy storage and V2G in a virtual power plant in a low-carbon urban park according to claim 1, characterized in that, The process of finding the optimal charging path for various types of users includes: With minimizing charging costs as the core decision-making objective, we construct a user profile for price-sensitive individuals. With the shortest total travel time as the core decision-making objective, construct a time-sensitive user profile; With the core decision-making objective of ensuring the remaining driving range after charging, we construct a user profile for those with range anxiety. Differentiated path utility functions are constructed for different types of users, and the dynamic Dijkstra algorithm is used to solve for the optimal charging path for each type of user in the charging network.

3. The method for coordinated scheduling of energy storage and V2G in a virtual power plant in a low-carbon urban park according to claim 1, characterized in that, The construction of the comprehensive satisfaction index specifically involves: using the extreme value normalization method to eliminate the dimensional differences between charging cost, time cost, and range anxiety, and constructing single-type satisfaction indices for each type of user; integrating the single-type satisfaction indices into a comprehensive satisfaction index through weighted summation; and constraining the user-side behavior boundaries that can participate in scheduling with the goal of maximizing the comprehensive satisfaction index.

4. The method for coordinated scheduling of energy storage and V2G in a virtual power plant in a low-carbon urban park according to claim 1, characterized in that, The differentiated behavioral constraints specifically include: determining the state-of-charge safety threshold, schedulable time periods, and adjustable power range for various types of users based on various user profiles.

5. The method for coordinated scheduling of energy storage and V2G in a virtual power plant in a low-carbon urban park according to claim 1, characterized in that, The specific steps for establishing the source-load-storage multi-resource coupling model are as follows: Based on differentiated behavioral constraints and actual equipment parameters in the park, mathematical models of various resources within the virtual power plant are established, including photovoltaic power output model, micro gas turbine operation model, building load model, electric vehicle charging and discharging model, and energy storage system operation model. Based on the two-way time-of-use pricing mechanism, this paper integrates the cost items of grid interaction, micro-turbine, energy storage, and photovoltaic curtailment, and adds the revenue from selling electricity to building loads and ancillary services to construct a comprehensive operating cost model for virtual power plants. The grid-connected power fluctuation rate is introduced as an auxiliary optimization target to complete the multi-resource coupling model of source-load-storage.

6. The method for coordinated scheduling of energy storage and V2G in a virtual power plant in a low-carbon urban park according to claim 1, characterized in that, The triggering logic and priority of the three energy storage-V2G collaborative operation modes are as follows: the backup capacity mode has the highest priority and is triggered first when the power shortage endangers the system safety; the fluctuation smoothing mode is the second highest and is triggered when the output fluctuation of renewable energy exceeds the set threshold; the peak shaving and valley filling mode has the lowest priority as the daily economic dispatch benchmark. All three modes must satisfy the reverse collaboration constraint and be arbitrated in order of priority.

7. The method for coordinated scheduling of energy storage and V2G in a virtual power plant in a low-carbon urban park according to claim 6, characterized in that, The triggering mechanism for the standby capacity mode is as follows: If the triggering conditions for the peak shaving and valley filling mode are not met, and the triggering conditions for the fluctuation smoothing mode are also not met, and the virtual power plant experiences a power shortage, then the reserve capacity mode is triggered. The triggering conditions for the peak shaving and valley filling mode are as follows: when the net load of the virtual power plant is greater than or equal to the peak threshold, the peak shaving phase is triggered, and energy storage and V2G are given priority to discharge; when the net load of the virtual power plant is less than the valley threshold, the valley filling phase is triggered, and energy storage and V2G are given priority to charge. The triggering condition for the fluctuation smoothing mode is as follows: if the triggering condition for the peak shaving and valley filling mode is not met, and the variance of the grid-connected power sliding window is detected to be greater than the set value or the rate of change of grid-connected power at adjacent times exceeds the ramp rate limit, then the fluctuation smoothing mode is triggered.

8. The method for coordinated scheduling of energy storage and V2G in a virtual power plant in a low-carbon urban park according to claim 1, characterized in that, The specific steps for defining the net load and applying reverse collaborative constraints are as follows: Define the net power of the energy storage system and the net power of the electric vehicle respectively, and apply the reverse cooperative constraint so that the signs of their net power are opposite at any scheduling time; The reverse coordination constraint includes a reverse coordination depth coefficient and a reverse coordination depth constraint. The reverse coordination depth coefficient is the ratio of the smaller of the absolute value of the net power of the energy storage system and the absolute value of the net power of the electric vehicle at the same time to the larger of the two values, and the denominator is avoided to be zero by a preset minimum positive number. The larger the value, the higher the degree of reverse matching between the power of the energy storage and the electric vehicle. The reverse coordination depth constraint is that the reverse coordination depth coefficient at any time is not lower than a preset reverse coordination depth threshold.

9. The method for coordinated scheduling of energy storage and V2G in a virtual power plant in a low-carbon urban park according to claim 1, characterized in that, The comprehensive objective function of the three-objective optimization scheduling model is a weighted fusion of the economic objective function value, the grid-connected load volatility objective function value, and the low-carbon objective function value. The economic objective function value is based on minimizing the full life cycle operating cost of the park, including grid interaction cost, micro gas turbine operating cost, energy storage system depreciation cost, photovoltaic curtailment penalty cost, and reverse collaboration penalty cost, while deducting building load electricity sales revenue and ancillary service revenue.

10. The method for coordinated scheduling of energy storage and V2G in a virtual power plant in a low-carbon urban park according to claim 9, characterized in that, The objective function value for the grid-connected load volatility is the variance between the grid-connected power and the average grid-connected power within the scheduling period.

Images