Method for participating in electric energy-frequency modulation market bidding by charging and replacing power station considering high-capacity power battery

Abstract

The invention discloses a method for participating in electric energy-frequency modulation market bidding by a charging station considering a high-capacity power battery, and relates to the field of electric power market bidding, and the method comprises the steps: building an equivalent energy storage aggregation model based on a high-capacity power battery cluster and a physical constraint framework of a pre-built charging station; generating a disordered battery replacement demand reference curve; combining the clearing price of the frequency modulation market to formulate a time-sharing power conversion price signal; establishing dual minimization targets with demand fluctuation and benefit loss as constraints, and solving a demand transfer result and standby battery resources by adopting a particle swarm algorithm; formulating a double-layer game strategy that the battery swap station participates in the electric energy and frequency modulation market; and converting the double-layer game strategy into a single-layer mixed integer linear programming model by utilizing an optimality condition and a large M method, and outputting an optimal comprehensive bidding method of the battery swap station by utilizing an alternating iteration algorithm. According to the method, the characteristics of the high-capacity battery are adapted, the aggregation model and the transportation logic are optimized, and the frequency modulation capacity and mileage income are improved.

Description

Technical Field

[0001] This invention relates to the field of electricity market bidding, and more specifically, to a method for charging and swapping stations with large-capacity power batteries to participate in electricity energy-frequency regulation market bidding. Background Technology

[0002] The large-scale grid connection of electric vehicles (EVs) has significantly increased the peak-shaving and frequency regulation pressure on the power grid, necessitating the use of vehicle-to-grid (V2G) technology to achieve a win-win situation for both EVs and the power grid. Battery swapping, as an important method for EV energy replenishment, can decouple the power battery from the EV in terms of time, space, and assets. Battery swapping stations (BSS), through centralized management of battery clusters, possess the potential for active load regulation and energy storage ancillary services, laying the foundation for their market-oriented operation.

[0003] However, existing technologies for battery swapping stations to participate in the electricity market have the following shortcomings: traditional technologies are mostly based on small-capacity battery designs, which are difficult to adapt to the aggregation characteristics of large-capacity power batteries. Small-capacity batteries require a large number of individual cells to meet the total capacity demand, resulting in complex charging and discharging control and delayed frequency regulation response. Meanwhile, the characteristics of large-capacity batteries have not been utilized, further exacerbating the mismatch between battery swapping demand and frequency regulation demand.

[0004] Insufficient coordination between battery swapping demand and market dispatch: Most studies set the battery swapping price as a constant value, failing to consider the guiding role of time-of-use battery swapping prices on EV users' battery swapping behavior. This results in the number and timing of redundant batteries at battery swapping stations not matching the demand of the frequency regulation market, thus limiting the revenue of frequency regulation services. Lack of coordination in dual-market bidding: Existing technologies mostly focus on battery swapping stations adjusting their charging and discharging strategies according to market prices, ignoring the impact of their bidding behavior on market prices; furthermore, they fail to effectively coordinate bidding decisions between the energy market and the frequency regulation market, leading to suboptimal revenue for battery swapping stations and difficulty in simultaneously meeting the peak shaving and frequency regulation needs of the power grid. Imbalance between economics and service guarantee: While some technologies propose balancing battery swapping services with power grid peak shaving, they lack a scientific revenue balancing mechanism, resulting in poor operational economics for battery swapping stations or the inability to guarantee the battery swapping needs of EV users.

[0005] Therefore, there is an urgent need for a battery swapping station price stabilization technology that can optimize redundant battery resources through demand response, coordinate bidding in both markets, and balance revenue and grid dispatch requirements.

[0006] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention

[0007] To address the problems in related technologies, this invention proposes a method for charging and swapping stations with large-capacity power batteries to participate in the bidding for electricity-frequency regulation market, thereby overcoming the aforementioned technical problems in existing related technologies.

[0008] Therefore, the specific technical solution adopted by the present invention is as follows:

[0009] This invention provides a method for charging and swapping stations considering large-capacity power batteries to participate in bidding in the electricity-frequency regulation market, including:

[0010] Based on the physical constraints of a large-capacity power battery cluster and a pre-established battery swapping station, an equivalent energy storage aggregation model is constructed. The probability of electric vehicle battery swapping is simulated through the equivalent energy storage aggregation model, and a baseline curve of disordered battery swapping demand is generated.

[0011] Based on the baseline curve of disordered battery swapping demand, combined with the frequency regulation market clearing price, a time-sharing battery swapping price signal is formulated to guide electric vehicle users to adjust their battery swapping time, so as to optimize the ideal quantity and time distribution of backup batteries in centralized charging stations.

[0012] Based on optimizing the ideal quantity and time distribution of backup batteries in centralized charging stations, a dual objective is established, and the particle swarm optimization algorithm is used to solve for the demand transfer results and backup battery resources.

[0013] Based on the demand transfer results and backup battery resources, a two-level game strategy for battery swapping stations to participate in the power and frequency regulation markets is formulated. By introducing multiple constraints, the coordinated allocation of resources and overall revenue of the two markets can be achieved.

[0014] By utilizing optimality conditions and the Big M method, the two-level game strategy is transformed into a solvable single-level mixed integer linear programming model. Then, using an alternating iterative algorithm, the optimal comprehensive bidding method for battery swapping stations is output.

[0015] Furthermore, based on the physical constraint framework of a large-capacity power battery cluster and pre-established battery swapping stations, an equivalent energy storage aggregation model is constructed; the equivalent energy storage aggregation model is used to simulate the probability of electric vehicle battery swapping and generate a baseline curve for disordered battery swapping demand, including:

[0016] A physical constraint framework is established with the two-layer architecture of the battery swapping station as the core, wherein the battery swapping station includes: a centralized charging station and a distributed distribution station;

[0017] By using distributed distribution stations for transportation, the batteries and electric vehicles can be decoupled in terms of time, space, and assets. The batteries can be replaced and charged at centralized charging stations and connected to the power grid.

[0018] During grid connection, an equivalent energy storage model is constructed by using an aggregation model and the Minkowski addition method to aggregate charging and discharging power and capacity.

[0019] The Monte Carlo simulation algorithm is used to simulate the probability distribution of electric vehicles' battery swapping behavior at different times of the day within a preset area, calculate the baseline value of battery swapping demand at different times under disordered conditions, and generate a baseline curve of disordered battery swapping demand.

[0020] Furthermore, during the grid connection process, an equivalent energy storage model is constructed using an aggregation model and the Minkowski addition method to aggregate charging and discharging power and capacity, including:

[0021] During grid connection, an aggregation model is used to reduce the difference between the centralized charging station and the replaced battery by using the grid connection status identifier variable, and the overall charging / discharging power of the centralized charging station is obtained by aggregation.

[0022] By introducing grid-connected state variables, the grid entry time and grid exit time of batteries entering / leaving centralized charging stations are defined. Combined with the charging and discharging characteristics of large-capacity batteries, the domain of a single battery model is extended to the entire scheduling cycle to fully depict the state switching process caused by transportation during the scheduling period.

[0023] Based on the full scheduling cycle and changes in battery grid connection status, the capacity envelope space constraint of the battery cluster is obtained through the Minkowski addition method. The equivalent energy storage model is obtained by comprehensively considering the impact of charging / discharging power, grid connection time / off-grid time, and changes in battery grid connection status on large-capacity batteries.

[0024] Furthermore, the method of simulating the probability distribution of electric vehicle battery swapping behavior at different times of the day within a preset area using the Monte Carlo simulation algorithm, calculating the baseline value of battery swapping demand at different times under disordered conditions, and generating the baseline curve of disordered battery swapping demand includes:

[0025] Using the Monte Carlo simulation algorithm, the initial battery level, daily mileage, and key parameters of the initial usage time of each electric vehicle in a preset area are randomly sampled, and the dynamic battery consumption of the electric vehicle during the entire day's driving process is simulated.

[0026] Based on the initial battery level and daily mileage, the initial vehicle usage period was determined by sampling.

[0027] Calculate the remaining driving range of the vehicle based on the initial battery level and power consumption parameters;

[0028] The actual mileage traveled from the start time period is obtained by summing the daily mileage and hourly mileage percentage functions.

[0029] When the actual driving mileage reaches the mileage that can continue to be driven, it is determined that the electric vehicle has undergone a battery swap during the current period;

[0030] By statistically analyzing the time distribution of battery swapping events for all electric vehicles, a baseline value for disordered battery swapping demand throughout the day is obtained, which characterizes the probability distribution features of battery swapping behavior within a preset area and generates a baseline curve for disordered battery swapping demand.

[0031] Furthermore, the step of formulating a time-of-use battery swapping price signal based on the disordered battery swapping demand baseline curve, combined with the frequency regulation market clearing price, and guiding electric vehicle users to adjust their battery swapping time to optimize the ideal quantity and time distribution of backup batteries in centralized charging stations includes:

[0032] Based on the differences in the clearing price of the frequency regulation market, the entire day is divided into high frequency regulation revenue period, medium frequency regulation revenue period and low frequency regulation revenue period, and corresponding differentiated time-of-use battery swapping prices are formulated.

[0033] Based on differentiated time-of-use battery swapping price signals, the aim is to guide electric vehicle users to shift their battery swapping demand from high-yield frequency regulation periods to low-yield periods, thereby obtaining the actual value of the shifted battery swapping demand and improving the availability of redundant batteries during high-yield frequency regulation periods.

[0034] By using the sensitivity coefficient of battery swapping price, the actual value of battery swapping demand after the transfer is optimized, and the ideal quantity and time distribution of backup batteries in centralized charging stations are obtained.

[0035] Furthermore, the process of establishing a dual objective of minimizing demand fluctuations and profit losses based on the adjusted battery swapping time, and using a particle swarm optimization algorithm to solve for the demand transfer results and backup battery resources, includes:

[0036] With the dual objectives of minimizing demand fluctuations and minimizing equivalent market benefit losses under time-of-use battery swapping pricing, a single objective function is integrated by weighting and summing the results using preset weighting coefficients.

[0037] The particle swarm optimization algorithm is used to solve the single objective function, and the outputs the demand transfer results and backup battery resources.

[0038] When the deviation of the demand transfer result and backup battery resources output by the particle swarm algorithm exceeds the tolerance, the outer adaptive boundary adjustment mechanism is introduced. In this case, the price search boundary is automatically scaled and the particle swarm algorithm search process is restarted.

[0039] Furthermore, the formulation of a two-tiered game strategy for battery swapping stations to participate in the electricity and frequency regulation markets based on demand transfer results and backup battery resources, and the introduction of multiple constraints to achieve coordinated allocation of resources and overall revenue management in both markets, includes:

[0040] The decision-making process for the upper-level battery swapping station is a bidding process aimed at maximizing the total revenue of the swapping station.

[0041] Lower-level market clearing refers to the joint clearing of the electricity market with the goal of minimizing total electricity purchase costs;

[0042] By introducing constraints on charging and discharging power, equivalent energy storage capacity, frequency regulation resources, and bidding prices, the decision-making process for upper-level battery swapping stations and the clearing process for lower-level markets can achieve coordinated allocation of resources and overall revenue sharing between the two markets.

[0043] Furthermore, the various constraints include: charging and discharging power constraints, equivalent energy storage capacity constraints, frequency regulation resource constraints, and bidding price constraints;

[0044] Among them, the charging and discharging power constraints are as follows: the charging and discharging power declared in each time period shall not exceed the power limit corresponding to its equivalent energy storage model, and the charging and discharging states are mutually exclusive.

[0045] Equivalent energy storage capacity constraints: The initial power evolution of equivalent energy storage in each time period must follow a continuous dynamic considering charging and discharging efficiency, and remain within the upper and lower limits of capacity.

[0046] Frequency regulation resource constraints: The declared frequency regulation capacity is limited by the total available capacity of the polymer battery, and the declared frequency regulation mileage is related to the frequency regulation capacity through a mileage coefficient;

[0047] Bidding price constraints: The declared price for electric energy charging and discharging, frequency regulation capacity, and mileage must be within the preset upper and lower limits of the market price.

[0048] Furthermore, the method of transforming the two-level game strategy into a solvable single-level mixed-integer linear programming model using optimality conditions and the Big M method, and outputting the optimal comprehensive bidding method for battery swapping stations using an alternating iterative algorithm, includes:

[0049] By utilizing optimality conditions, the lower-level market clearing is transformed into an equivalent optimality condition constraint, thus decomposing the two-level game strategy into a single-level optimization problem.

[0050] By using the Big M method and the strong duality theorem, the nonlinear complementary relaxation constraints in the single-layer optimization problem are linearized to obtain a solvable single-layer mixed integer linear programming model, so as to achieve a unified solution for sequential / joint clearing scenarios.

[0051] Using an alternating iterative algorithm, the price signal is updated with the market clearing result, and the demand transfer result and backup battery resources are fed back in to output the optimal comprehensive bidding method for battery swapping stations.

[0052] Furthermore, the method of using an alternating iterative algorithm to update price signals with market clearing results, replenish demand transfer results and backup battery resources, and output the optimal comprehensive bidding method for battery swapping stations includes:

[0053] Initialize the baseline curve of disordered battery swapping demand, the electricity and frequency regulation market and the frequency regulation market clearing price, and generate an initial population of time-of-use battery swapping prices;

[0054] The particle swarm optimization algorithm is used to solve the initial population of time-of-use battery swapping prices, and the battery swapping demand and backup battery time distribution after demand shift are obtained.

[0055] Based on backup battery resources, a single-layer mixed integer linear programming model is used to solve the problem and obtain the optimal bidding strategy and market clearing price for battery swapping stations.

[0056] When the rate of change in battery swapping service revenue is greater than one percent, the price inputs of the battery swapping demand after the demand shift and the time distribution of backup batteries are updated with the new market clearing price, and the particle swarm algorithm and single-layer mixed integer linear programming model are repeatedly used to solve the problem.

[0057] When the rate of change in revenue from battery swapping services is less than one percent, or when the number of iterations reaches a preset upper limit, the iteration stops, and the optimal comprehensive bidding method for battery swapping stations is output.

[0058] The beneficial effects of this invention are as follows:

[0059] 1) This invention adapts to the characteristics of large-capacity batteries, optimizes the aggregation model and transportation logic, improves frequency regulation capacity and mileage benefits, while reducing unit charging costs and control complexity. It also conducts exploratory experiments on market impact and achieves the optimal bidding strategy for battery swapping stations.

[0060] 2) Under joint clearing, this invention prioritizes redundant batteries to serve periods of high frequency regulation demand, increasing the winning bids for frequency regulation capacity and mileage while reducing the corresponding clearing price. Under sequential clearing, energy is prioritized, and the available capacity for frequency regulation is limited. Overall, the strategy described in this paper can reduce charging costs, significantly improve frequency regulation revenue, and maintain the revenue from battery swapping services unchanged. Attached Figure Description

[0061] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0062] Figure 1 This is a flowchart of a method for charging and swapping stations with large-capacity power batteries to participate in the bidding for electricity-frequency regulation market according to an embodiment of the present invention;

[0063] Figure 2 This is a flowchart of step S1 in the method for charging and swapping stations of large-capacity power batteries to participate in the bidding for electricity-frequency regulation market according to an embodiment of the present invention;

[0064] Figure 3 This is a flowchart of step S2 in the method for charging and swapping stations of large-capacity power batteries to participate in the bidding for electricity-frequency regulation market according to an embodiment of the present invention;

[0065] Figure 4This is a flowchart of step S3 in the method for charging and swapping stations of large-capacity power batteries to participate in the bidding for electricity-frequency regulation market according to an embodiment of the present invention;

[0066] Figure 5 This is a flowchart of step S5 in the method for charging and swapping stations of large-capacity power batteries to participate in the bidding for electricity-frequency regulation market according to an embodiment of the present invention. Detailed Implementation

[0067] To further illustrate the various embodiments, the present invention provides accompanying drawings, which are part of the disclosure of the present invention. These drawings are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these drawings, those skilled in the art should be able to understand other possible implementation methods and the advantages of the present invention.

[0068] According to embodiments of the present invention, a method is provided for charging and swapping stations that take into account large-capacity power batteries to participate in the bidding for electricity-frequency regulation market.

[0069] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figure 1 As shown, the method for charging and swapping stations considering large-capacity power batteries to participate in the electricity-frequency regulation market bidding according to an embodiment of the present invention includes:

[0070] Step S1: Establish the physical constraint framework of the battery swapping station. Based on the large-capacity power battery cluster, construct an equivalent energy storage aggregation model and simulate the probability of electric vehicle (EV) battery swapping to generate a baseline curve for disordered battery swapping demand.

[0071] In this optional embodiment, the construction of an equivalent energy storage aggregation model based on a large-capacity power battery cluster and a pre-established physical constraint framework of battery swapping stations; and the simulation of the probability of electric vehicle battery swapping through the equivalent energy storage aggregation model to generate a baseline curve for disordered battery swapping demand, including:

[0072] A physical constraint framework is established with the two-layer architecture of the battery swapping station as the core, wherein the battery swapping station includes: a centralized charging station (CCS) and a distributed distribution station (DS).

[0073] By using distributed distribution stations for transportation, the batteries and electric vehicles can be decoupled in terms of time, space, and assets. The batteries can be replaced and charged at centralized charging stations and connected to the power grid.

[0074] During grid connection, an equivalent energy storage model is constructed by using an aggregation model and the Minkowski addition method to aggregate charging and discharging power and capacity.

[0075] The Monte Carlo simulation algorithm is used to simulate the probability distribution of electric vehicles' battery swapping behavior at different times of the day within a preset area, calculate the baseline value of battery swapping demand at different times under disordered conditions, and generate a baseline curve of disordered battery swapping demand.

[0076] In this optional embodiment, the step of constructing an equivalent energy storage model through aggregation model and Minkowski addition during grid connection to aggregate charging and discharging power and capacity includes:

[0077] During grid connection, an aggregation model is used to reduce the difference between the centralized charging station and the replaced battery by using the grid connection status identifier variable, and the overall charging / discharging power of the centralized charging station is obtained by aggregation.

[0078] By introducing grid-connected state variables, the grid entry time and grid exit time of batteries entering / leaving centralized charging stations are defined. Combined with the charging and discharging characteristics of large-capacity batteries, the domain of a single battery model is extended to the entire scheduling cycle to fully depict the state switching process caused by transportation during the scheduling period.

[0079] Based on the full scheduling cycle and changes in battery grid connection status, the capacity envelope space constraint of the battery cluster is obtained through the Minkowski addition method. The equivalent energy storage model is obtained by comprehensively considering the impact of charging / discharging power, grid connection time / off-grid time, and changes in battery grid connection status on large-capacity batteries.

[0080] In this optional embodiment, the step of simulating the probability distribution of electric vehicle battery swapping behavior at different times of the day within a preset area using a Monte Carlo simulation algorithm, calculating the baseline value of battery swapping demand at each time period under disordered conditions, and generating a baseline curve for disordered battery swapping demand includes:

[0081] Using the Monte Carlo simulation algorithm, key parameters such as the initial state of charge (SOC), daily mileage, and initial usage time of each electric vehicle within a preset area are randomly sampled, and the dynamic energy consumption of electric vehicles during the entire day's driving is simulated. Based on the initial SOC and daily mileage, the initial usage time is determined by sampling. According to the initial SOC and energy consumption parameters, the mileage that the vehicle can continue to drive is calculated. Based on the sum of the daily mileage and the hourly mileage percentage function, the actual mileage from the initial time period is obtained. When the actual mileage reaches the mileage that can continue to drive, it is determined that the electric vehicle has engaged in battery swapping behavior in the current time period. By statistically analyzing the time distribution of battery swapping events of all electric vehicles, the baseline value of disordered battery swapping demand for the entire day is obtained, which characterizes the probability distribution characteristics of battery swapping behavior within the preset area and generates a baseline curve of disordered battery swapping demand.

[0082] Specifically, such as Figure 2As shown, a unified "station-grid-market" physical constraint framework is first established for battery swapping stations to participate in the electricity-frequency regulation market, ensuring that subsequent strategies are optimized within operable, accessible, and measurable boundaries. Battery swapping stations are divided into centralized charging stations and distributed distribution stations. Distributed distribution stations facilitate transportation, decoupling the time, space, and assets of batteries and vehicles. Swapped batteries are then uniformly charged and connected to the grid at centralized charging stations. First, a single battery model for a battery swapping station is constructed, extending its domain to the entire scheduling cycle. By introducing aggregate variables and utilizing the Minkowski addition method, an "equivalent energy storage" model is built, aggregating battery charging and discharging power and capacity.

[0083] Compared to small-capacity batteries, large-capacity power batteries can have a single cell capacity of over 80kWh. This allows for a reduction in the number of batteries during aggregation (reducing the number of cells by 40%-60% for the same total capacity), lowering the complexity of charge and discharge control and minimizing the impact of capacity decay on model accuracy. The system explicitly considers the differences in grid connection / disconnection timing caused by transportation delays, forming a cross-time-period grid connection status and capacity evolution. The upper limit of total charge / discharge power, charge / discharge efficiency, SOC upper and lower limits, capacity jumps caused by grid connection status switching, and single-time period length are uniformly set; ensuring that both electrical energy and frequency regulation services are scheduled within the same physical boundary. The system uses Monte Carlo simulation to obtain the disordered battery swapping demand curve as the base load. Based on the characteristic that a single recharge of a large-capacity battery can meet the EV's 300-500km range, the battery swapping interval is extended from the traditional 1-2 days for small-capacity batteries to 3-5 days during simulation, making the demand curve more closely match actual user behavior and reducing the calculation deviation of the base load. Furthermore, time-sharing battery swapping pricing triggers "transferable demand." Traditional battery generator sets adhere to capacity limits and clearing constraints; battery swapping stations, using battery clusters as rapidly adjustable resources, are constrained by aggregated battery capacity and transportation timing, making them more suitable for rapid response and mileage-based frequency regulation. Based on a two-tier architecture of centralized charging centers and distributed battery swapping service stations, a unified physical constraint system is established, clarifying the mathematical boundaries of battery flow, charging / discharging, and grid-connected scheduling. The core formulas and logic adjustments are as follows: The CCS battery inventory time distribution model considers the battery transportation time from DSS to CCS in different regions. CCS during the time period The battery inventory is determined by the previous period's inventory, historical battery swapping demand, and future battery swapping demand. A battery inventory calculation model is designed as follows:

[0084] ;

[0085] In the formula, This indicates the actual number of batteries in CCS during time period τ; Indicates time period Total demand for EV battery swapping; This represents the battery transportation time from DSS to CCS in different regions. This formula accurately reflects the impact of transportation delay on the distribution of battery grid connection time.

[0086] 1. Battery cluster charging and discharging power aggregation model at battery swapping stations: Due to the differences in the grid connection time of batteries within the station, a grid connection status identifier variable is introduced. (A value of 1 indicates that battery n is in the time period) (Grid-connected, with a value of 0 indicating off-grid operation), the total charging and discharging power of the battery swapping station is aggregated to obtain the overall charging and discharging power of the station:

[0087] ;

[0088] In the formula, , These represent the time periods of the battery swapping station. Total charging power and total discharging power; This represents the collection of all batteries within the battery swapping station; , These represent the time periods of a single battery n. Charging power and discharging power; This variable represents the grid connection status identifier.

[0089] 2. Equivalent energy storage characterization model for battery clusters:

[0090] To expand the domain of individual scheduling for large-capacity power batteries within a battery swapping station, grid-connected state variables are introduced. (A value of 1 indicates that battery n is in the time period) Grid connection (a value of 0 indicates grid disconnection) is defined as the grid connection period when battery n enters the centralized charging station (CCS), and the grid disconnection period when battery n leaves the CCS.

[0091] ;

[0092] Based on the charging and discharging characteristics of large-capacity batteries, the domain of the single-cell charging and discharging model is extended to the entire scheduling cycle, as shown in the equation:

[0093] ;

[0094] In the formula, Indicates that battery n is in the time period The amount of electricity; , These represent the time periods of battery n. Charging power and discharging power; Let n represent the set of grid connection times for battery n. , These represent charging efficiency and discharging efficiency, respectively. Indicates the length of a single time period.

[0095] Based on the different grid connection status of battery n, further... The changes are divided into three situations: the battery grid connection period, the grid connection duration, and the off-grid period, as shown in the formula:

[0096] ;

[0097] In the formula, Indicates that battery n is in the time period The amount of electricity; This indicates the grid connection capacity of battery n (which must be fully charged to provide battery swapping services). This represents the off-grid capacity of battery n; , These represent charging efficiency and discharging efficiency, respectively. Indicates the length of a single time period; , These represent the time periods of a single battery n. Charging power and discharging power; Equivalent to (Indicates time period) (Battery n connected to the network). Equivalent to (Indicates time period) (battery n is off-grid), based on which this formula can be integrated into a unified form:

[0098] ;

[0099] The capacity envelope space constraint of the battery cluster is obtained by processing the above two equations using the Minkowski addition method:

[0100] ;

[0101] Introducing the equivalent energy storage variable of the battery swapping station (total charging power) Total discharge power Equivalent energy storage capacity The correspondence between this and single-cell parameters, and the definition of equivalent energy storage parameters. Further, this is combined with CCS time periods. The total number of grid-connected batteries, and the time period The number of batteries connected to the grid equals The number of off-grid batteries equals Based on the relationship, this formula is simplified into an equivalent energy storage parameter expression adapted to the battery swapping demand, ultimately yielding a complete constraint model for the equivalent energy storage of a battery swapping station:

[0102] ;

[0103] In the formula, This refers to the collection of large-capacity power batteries within a battery swapping station. , These represent the maximum charging power and maximum discharging power of a single high-capacity battery, respectively. , These represent the upper and lower limits of the capacity of a single high-capacity battery, respectively. , These represent the maximum charging power and maximum discharging power of the equivalent energy storage at the battery swapping station, respectively. , These represent the upper and lower limits of the equivalent energy storage capacity, respectively. This represents the equivalent energy storage capacity jump caused by changes in the battery's grid connection status. This indicates the transportation time of the battery between the CCS and the distributed distribution station (DS). Indicates time period The battery swapping demand. Considering the impact of changes in charging and discharging power, charging and discharging efficiency, and battery grid connection status on capacity, the equivalent energy storage capacity of a battery swapping station must meet the following constraints:

[0104] ;

[0105] In the formula, , These represent the time periods of the battery swapping station. Maximum charging power and maximum discharging power; Indicates equivalent energy storage capacity; , These represent charging efficiency and discharging efficiency, respectively. This indicates the equivalent energy storage capacity jump caused by changes in the battery's grid connection status. , These represent the battery's capacity when connected to the grid and its capacity when disconnected from the grid, respectively. This model provides a unified physical boundary for the coordinated scheduling of power services and frequency regulation services.

[0106] 3. Generation of Disordered Battery Swapping Demand Curve: Using the Monte Carlo simulation method, the probability distribution of EV battery swapping behavior occurring at different times throughout the day in the simulated area is obtained, and the baseline value of battery swapping demand at each time period under disordered conditions is calculated. This serves as the foundational load data for subsequent optimization of battery swapping demand. Specifically, Monte Carlo simulation technology is used to simulate the dynamic energy consumption of each electric vehicle throughout the day by randomly sampling key parameters such as its initial state of charge (SOC), daily mileage, and initial usage time within a random sampling area. The initial SOC and daily mileage are generated based on a preset distribution, and the initial usage time is determined through sampling. The following two equations are then combined to calculate the vehicle's remaining mileage based on the initial SOC and energy consumption parameters. ,

[0107] ;

[0108] ;

[0109] In the formula, and Represent the total mileage of the k-th EV and the mileage at the time of travel, respectively. Remaining battery power at a given time period; This indicates the electricity consumption of an EV per 100km of driving. Indicates the critical charge level for battery swapping; This represents the total mileage driven by EVk in one day; This indicates the percentage of EVk's mileage per hour out of the total mileage.

[0110] This is the critical battery level for battery swapping. Simultaneously, the actual mileage since the start of the period is calculated by summing daily mileage and hourly mileage percentages using a function. When the actual mileage reaches... When the vehicle is in use, it is determined that it has engaged in battery swapping during the current time period. By statistically analyzing the time distribution of battery swapping events for all electric vehicles, the baseline value of disordered battery swapping demand for 24 time periods throughout the day is obtained, thereby effectively characterizing the probability distribution features of battery swapping behavior in the region.

[0111] Step S2: Based on the baseline curve of disordered battery swapping demand and combined with the frequency regulation market clearing price, formulate a time-sharing battery swapping price signal and guide electric vehicle users to adjust their battery swapping time in order to optimize the ideal quantity and time distribution of backup batteries in centralized charging stations.

[0112] In this optional embodiment, the step of formulating a time-of-use battery swapping price signal based on the disordered battery swapping demand baseline curve and in conjunction with the frequency regulation market clearing price, and guiding electric vehicle users to adjust their battery swapping time to optimize the ideal quantity and time distribution of backup batteries in centralized charging stations includes:

[0113] Based on the differences in the clearing price of the frequency regulation market, the entire day is divided into high frequency regulation revenue period, medium frequency regulation revenue period and low frequency regulation revenue period, and corresponding differentiated time-of-use battery swapping prices are formulated.

[0114] Based on differentiated time-of-use battery swapping price signals, the aim is to guide electric vehicle users to shift their battery swapping demand from high-yield frequency regulation periods to low-yield periods, thereby obtaining the actual value of the shifted battery swapping demand and improving the availability of redundant batteries during high-yield frequency regulation periods.

[0115] By using the sensitivity coefficient of battery swapping price, the actual value of battery swapping demand after the transfer is optimized, and the ideal quantity and time distribution of backup batteries in centralized charging stations are obtained.

[0116] Specifically, such as Figure 3As shown, physical scheduling capabilities are coupled with market incentives. Specifically, time-of-use battery swapping prices are set based on market clearing prices, EV clusters are scheduled to adjust battery swapping behavior, and battery charging and discharging strategies at battery swapping stations are optimized. This results in the distribution of redundant batteries in terms of quantity and timing that can meet market demand, constructing a revenue accounting system based on "electric energy + frequency regulation capacity / mileage" to form comparable economic objectives. Using the node marginal electricity price as a benchmark, charging incurs electricity purchase costs, while discharging generates electricity sales revenue, with the net amount entering the total revenue function. A dual payment structure of capacity (Cap) and mileage (mile) is adopted, allowing battery swapping stations to receive frequency regulation revenue based on declared capacity and actual mileage; the mileage multiplier reflects the amplification / reduction of mileage revenue due to the dynamic tracking performance of resources. Setting time-of-use battery swapping prices based on the frequency regulation clearing price guides users to "transfer battery swapping demand from high-return periods to low-return periods," improving the availability of redundant batteries during periods of high frequency regulation demand. The system introduces a dual objective of minimizing demand fluctuations and minimizing equivalent market profit losses under time-sharing battery swapping pricing. This aims to both smooth out net battery swapping load and control market opportunity losses caused by relinquishment. Furthermore, it uses the constraint of "unchanged total service revenue from battery swapping" to prevent price manipulation.

[0117] Based on the market clearing price of frequency regulation, time-based battery swapping pricing is established to guide EV users to adjust their battery swapping time and optimize the quantity and time distribution of backup batteries. The core formula and logic are adjusted as follows:

[0118] 1. Calculation of market opportunity loss for single batteries: Time period The formula for calculating the potential gains (i.e., market opportunity loss) of batteries used to meet battery swapping needs, by forgoing participation in the frequency regulation market, is adjusted as follows:

[0119] ;

[0120] In the formula, Indicates time period Loss of market opportunities for individual batteries This indicates the capacity-clearing electricity price in the frequency regulation market. This formula represents the rated capacity of a single battery and provides an economic basis for time-of-use battery swapping pricing.

[0121] 2. Time-of-use pricing and demand shift model for battery swapping: Based on the difference in the clearing price of the frequency regulation market, a day is divided into high-profit periods for frequency regulation. Mid-yield period Low-yield periods Correspondingly, time-based battery swapping prices will be set:

[0122] ;

[0123] Time-of-use battery swapping prices and benchmark price The difference will trigger a shift in EV users' battery swapping needs. The formula for calculating the shift is:

[0124] ;

[0125] The final actual value of battery swapping demand after demand shift is obtained:

[0126] ;

[0127] In the formula, , , These represent the battery swapping prices for high, medium, and low-yield periods, respectively. This represents the sensitivity coefficient of battery swapping demand to price (a negative value indicates that demand decreases when prices rise). This model represents the battery swapping demand after the demand shift, and it demonstrates the guiding role of price on battery swapping behavior.

[0128] 3. Multi-objective optimization and constraints: Minimizing battery swapping demand fluctuations:

[0129] ;

[0130] In the formula, This indicates the demand for battery swapping after the demand shift. ( (This is the average value of the battery swapping demand in each time period after the demand shift), used to mitigate the impact of battery swapping load on the power grid;

[0131] The objective is to minimize market opportunity loss.

[0132] ;

[0133] Used to control frequency regulation revenue loss due to demand shift; stability constraint on total revenue of battery swapping services:

[0134] ;

[0135] To prevent battery swapping stations from manipulating prices to gain excessive profits and to protect the rights and interests of EV users; time-of-day price limits and upper / lower limits:

[0136] ;

[0137] In the formula, , These represent the lower and upper limits of the battery swapping price during high-yield periods, respectively. , These represent the lower and upper limits of battery swapping prices during periods of low profitability.

[0138] Step S3: Based on the adjusted battery swapping time, establish a dual objective of minimization constrained by demand fluctuations and profit losses, and use the particle swarm optimization algorithm to solve for the demand transfer results and backup battery resources.

[0139] In this optional embodiment, the step of establishing a minimum dual objective based on the adjusted battery swapping time, constrained by demand fluctuations and profit losses, and using a particle swarm optimization algorithm to solve for the demand transfer results and backup battery resources includes:

[0140] With the dual objectives of minimizing demand fluctuations and minimizing equivalent market benefit losses under time-of-use battery swapping pricing, a single objective function is integrated by weighting and summing the results using preset weighting coefficients.

[0141] The single objective function is solved using the particle swarm optimization (PSO) algorithm, and the demand transfer results and backup battery resources are output.

[0142] When the deviation of the demand transfer result and backup battery resources output by the particle swarm algorithm exceeds the tolerance, the outer adaptive boundary adjustment mechanism is introduced. In this case, the price search boundary is automatically scaled and the particle swarm algorithm search process is restarted.

[0143] Specifically, such as Figure 4 As shown, the model solution method is as follows: the dual objectives of "minimizing fluctuations in battery swapping demand" and "minimizing market opportunity loss" are combined using weighting coefficients. (satisfy Integrate into a single objective function The Particle Swarm Optimization (PSO) algorithm is used to solve this single-objective problem, outputting the optimal time-of-use battery swapping price and the corresponding backup battery time distribution. The optimization problem is modeled as a single-objective function with three variables (peak, flat, and valley electricity prices), and the objective function comprehensively considers demand equilibrium (…). Item) and system benefits ( The weighted sum of (items), where, This method smooths the load curve by minimizing the variance of battery swapping demand in different time periods. The goal is to maximize the total system revenue. During the initialization phase of the PSO algorithm, a population of multiple particles is randomly generated, with each particle representing a set of candidate electricity price solutions, and these solutions are randomly distributed within a preset electricity price boundary.

[0144] During the iterative optimization process, each particle dynamically updates its velocity and position based on its individual historical best position and the global best position of the group. Specifically, the velocity update formula comprehensively considers inertia, individual cognition, and social cognition, while the position update ensures that it does not exceed the feasible solution space. After each iteration, the fitness values ​​(i.e., objective function values) of all particles are re-evaluated, and the individual and global best solutions are updated.

[0145] To enhance the algorithm's practicality, this method also incorporates an outer adaptive boundary adjustment mechanism: when the deviation between the optimal solution obtained by PSO and the benchmark return exceeds the tolerance threshold, the search boundary is dynamically scaled according to the direction of the deviation—the boundary is lowered overall when the return is too high, and moved upwards when the return is insufficient, and the PSO search is restarted within the new boundary. This dynamic adjustment strategy ensures convergence accuracy while avoiding search failures caused by improper initial boundary settings.

[0146] Finally, when the relative deviation of revenue meets the convergence condition or reaches the maximum number of outer iterations, the algorithm outputs the optimal electricity price scheme that makes the system revenue close to the benchmark value and the demand distribution relatively balanced, providing a reasonable electricity price signal benchmark for subsequent stages.

[0147] Step S4: Based on the demand transfer results and backup battery resources, formulate a two-level game strategy for the battery swapping station to participate in the power and frequency regulation markets, and introduce multiple constraints to achieve coordinated allocation of resources and overall revenue in both markets.

[0148] In this optional embodiment, the step of formulating a two-tiered game strategy for the battery swapping station to participate in the electricity and frequency regulation markets based on demand transfer results and backup battery resources, and introducing multiple constraints to achieve coordinated allocation of resources and overall revenue management in both markets, includes:

[0149] The decision-making process for the upper-level battery swapping station is a bidding process aimed at maximizing the total revenue of the swapping station.

[0150] Lower-level market clearing refers to the joint clearing of the electricity market with the goal of minimizing total electricity purchase costs;

[0151] By introducing constraints on charging and discharging power, equivalent energy storage capacity, frequency regulation resources, and bidding prices, the decision-making process for upper-level battery swapping stations and the clearing process for lower-level markets can achieve coordinated allocation of resources and overall revenue sharing between the two markets.

[0152] In this optional embodiment, the multiple constraints include: charging and discharging power constraints, equivalent energy storage capacity constraints, frequency regulation resource constraints, and bidding price constraints;

[0153] Among them, the charging and discharging power constraints are as follows: the charging and discharging power declared in each time period shall not exceed the power limit corresponding to its equivalent energy storage model, and the charging and discharging states are mutually exclusive.

[0154] Equivalent energy storage capacity constraints: The initial power evolution of equivalent energy storage in each time period must follow a continuous dynamic considering charging and discharging efficiency, and remain within the upper and lower limits of capacity.

[0155] Frequency regulation resource constraints: The declared frequency regulation capacity is limited by the total available capacity of the polymer battery, and the declared frequency regulation mileage is related to the frequency regulation capacity through a mileage coefficient;

[0156] Bidding price constraints: The declared price for electric energy charging and discharging, frequency regulation capacity, and mileage must be within the preset upper and lower limits of the market price.

[0157] Specifically, the continuous interaction between the two markets is abstracted as a two-layer game of "optimal bidding for battery swapping stations - optimal market clearing," and two mechanisms, sequential clearing and joint clearing, are given. The solution is ultimately found using a standardized optimization framework. Upper-layer (battery swapping station) decision-making: Declaring charging / discharging power and price, frequency regulation capacity and mileage, and their prices, with the goal of maximizing total revenue; prioritizing the fulfillment of battery swapping service capacity constraints. Lower-layer (market) clearing: Joint clearing: Merging and optimizing electrical energy and frequency regulation, with the goal of minimizing total system cost while simultaneously determining the winning bid quantity and price in both markets. The upper and lower layers share physical constraints (power / capacity / SOC / timing), bidding upper and lower limits, and market demand constraints; the clearing price corresponds to dual variables (node ​​marginal electricity price and frequency regulation capacity / mileage electricity price). The "post-response demand curve and redundant battery timing" given in the first stage serve as the available resource boundary in the second stage, thus explicitly incorporating price-driven behavioral changes into the bidding and clearing process.

[0158] Based on the demand transfer results and backup battery resources obtained in the first phase, a collaborative bidding strategy for battery swapping stations to participate in both markets is formulated, clarifying the interactive mechanism of "battery swapping station bidding - market clearing". The core formula and logic are adjusted as follows:

[0159] 1. Total Revenue Objective Function for Battery Swapping Stations: The objective is to maximize the total operating revenue of battery swapping stations. This revenue comprises revenue from the electricity market, frequency regulation market, and battery swapping services. The formula is adjusted as follows:

[0160] ;

[0161] In the formula, This represents the total revenue of the battery swapping station; Indicates revenue from the electricity market; This indicates that the electricity market has cleared its electricity price. This indicates revenue from the FM market; , These represent the bid amount for frequency modulation capacity and the bid amount for frequency modulation mileage, respectively. This indicates the clearing price for frequency regulation mileage; This indicates the revenue generated from battery swapping services.

[0162] 2. Bidding constraints for battery swapping stations: Charging and discharging power constraints:

[0163] ;

[0164] In the formula, , These represent the time periods of the battery swapping station. The declared charging power and discharging power , These represent the charging status indicator and the discharging status indicator (with values ​​of 0 or 1), respectively.

[0165] Equivalent energy storage capacity constraints:

[0166] ;

[0167] In the formula, and These represent the charging and discharging efficiencies of the equivalent energy storage, respectively.

[0168] Frequency modulation resource constraints:

[0169] ;

[0170] In the formula, , These represent the declared frequency modulation capacity and frequency modulation mileage, respectively. This indicates the maximum frequency modulation capacity that can be declared; This indicates the frequency regulation mileage coefficient of the battery swapping station. The maximum declared frequency regulation capacity of the battery swapping station directly depends on the total aggregate capacity of the large-capacity batteries. For example, 100 80kWh large-capacity batteries can provide 8MWh of frequency regulation capacity, which is 100% higher than the same number of 40kWh batteries. This significantly enhances the competitiveness of the battery swapping station in the frequency regulation market and makes it easier to win bids for high mileage.

[0171] Bidding price constraints:

[0172] ;

[0173] In the formula, , , These represent the declared electricity market charging and discharging prices, frequency regulation capacity prices, and frequency regulation mileage prices, respectively, with the corresponding min and max subscript parameters being the upper and lower limits of each price.

[0174] 3. Electricity Market Clearing Model: Joint Clearing Mechanism: Integrating the clearing processes of the electricity market and the frequency regulation market, with the goal of minimizing the total electricity purchase cost of "electricity + frequency regulation".

[0175] ;

[0176] ;

[0177] ;

[0178] In the formula, This represents the total cost of electricity purchase; , These represent the bid amount for frequency modulation capacity and the bid amount for frequency modulation mileage, respectively. , These represent the time periods of the battery swapping station. Total charging power and total discharging power; , These represent the declared frequency modulation capacity and frequency modulation mileage, respectively. Indicates time period The charging and discharging prices declared by battery swapping stations in the electricity market. This represents a collection of conventional generator sets. This indicates a collection of professional frequency regulation units. This indicates the cost of purchasing electricity in the electricity market. This indicates the cost of purchasing electricity in the frequency regulation market. and Indicates time period The amount of electricity generated by conventional generator unit m in the electricity market and the price it bids for. This indicates the capacity price declared by frequency regulation unit i; additionally, a "coordinated constraint between charging / discharging power and frequency regulation capacity" is added to achieve coordinated and optimized allocation of resources from both markets.

[0179] ;

[0180] ;

[0181] After aggregation, the total charging and discharging power limit of the battery swapping station is increased, which can quickly respond to the power regulation needs of the power market during joint clearing, while reserving more capacity for frequency regulation services, reducing conflicts between resources in the two markets, and improving collaborative optimization efficiency.

[0182] Step S5: Using the optimality condition (KKT) and the Big M method, the two-level game strategy is transformed into a solvable single-level mixed integer linear programming model (MILP), and the optimal comprehensive bidding method (final bidding strategy) for the battery swapping station is output using the alternating iterative algorithm.

[0183] In this optional embodiment, the method of transforming the two-level game strategy into a solvable single-level mixed-integer linear programming model using optimality conditions and the Big M method, and outputting the optimal comprehensive bidding method for the battery swapping station using an alternating iterative algorithm, includes:

[0184] By utilizing optimality conditions, the lower-level market clearing is transformed into an equivalent optimality condition constraint, thus decomposing the two-level game strategy into a single-level optimization problem.

[0185] By using the Big M method and the strong duality theorem, the nonlinear complementary relaxation constraints in the single-layer optimization problem are linearized to obtain a solvable single-layer mixed integer linear programming model, so as to achieve a unified solution for sequential / joint clearing scenarios.

[0186] Using an alternating iterative algorithm, the price signal is updated with the market clearing result, and the demand transfer result and backup battery resources are fed back in to output the optimal comprehensive bidding method for battery swapping stations.

[0187] In this optional embodiment, the method of using an alternating iterative algorithm to update price signals with market clearing results, reinject demand transfer results and backup battery resources, and output the optimal comprehensive bidding method for battery swapping stations includes:

[0188] Initialize the baseline curve of disordered battery swapping demand, the electricity and frequency regulation market and the frequency regulation market clearing price, and generate an initial population of time-of-use battery swapping prices;

[0189] The particle swarm optimization algorithm is used to solve the initial population of time-of-use battery swapping prices, and the battery swapping demand and backup battery time distribution after demand shift are obtained.

[0190] Based on backup battery resources, a single-layer mixed integer linear programming model is used to solve the problem and obtain the optimal bidding strategy and market clearing price for battery swapping stations.

[0191] When the rate of change in battery swapping service revenue is greater than one percent, the price inputs of the battery swapping demand after the demand shift and the time distribution of backup batteries are updated with the new market clearing price, and the particle swarm algorithm and single-layer mixed integer linear programming model are repeatedly used to solve the problem.

[0192] When the rate of change in revenue from battery swapping services is less than one percent, or when the number of iterations reaches a preset upper limit, the iteration stops, and the optimal comprehensive bidding method for battery swapping stations is output.

[0193] Specifically, such as Figure 5 As shown, based on the aforementioned constraints and objectives, a feasible solution process of "first-stage multi-objective - second-stage mixed integer linearization" is adopted to ensure that the calculation is reproducible and the strategy is convergent.

[0194] Phase 1 (Demand Response Optimization): The objectives of "minimizing volatility" and "minimizing equivalent market loss" are transformed into a single objective through weighted summation. The Particle Swarm Optimization (PSO) algorithm is used to generate time-of-use battery swapping prices, and the demand and redundancy timing after the response are obtained.

[0195] The second stage (integrated bidding-clearing): The two-level game is transformed into a single-level problem using KKT conditions and strong duality. The complementary relaxation constraints are linearized using the Big M method. By introducing Boolean variables and multiplying them by a sufficiently large positive number M, the complementary relaxation constraints are expressed as linearized constraints, resulting in a usable mixed linear programming (MILP) model, which enables a unified solution for sequential / joint clearing scenarios.

[0196] Alternating Iteration: Price signals are updated based on market clearing results, and the demand and redundant configurations from the first phase are reinjected; convergence occurs when the change in battery swapping service revenue is less than a threshold (e.g., 1%) or the iteration limit is reached. Under joint clearing, redundant batteries prioritize serving periods of high frequency regulation demand, increasing the winning bids for frequency regulation capacity and mileage while reducing the corresponding clearing price; under sequential clearing, energy is prioritized, and available frequency regulation capacity is limited. Overall, the strategy presented in this paper can reduce charging costs, significantly improve frequency regulation revenue, while maintaining unchanged battery swapping service revenue.

[0197] By iteratively applying a two-stage model, the dynamic coupling between the shift in battery swapping demand and the bidding process in both markets is achieved, ensuring strategy convergence. The core logic is adjusted as follows:

[0198] 1. Model Transformation and Solution Techniques: In response to the two-layer game characteristics of the second stage "bidding for battery swapping stations - market clearing", the KKT (Karush-Kuhn-Tucker) optimality condition is adopted to transform the lower-level market clearing model into an equivalent KKT constraint, thus decomposing the two-layer optimization problem into a single-layer optimization problem. Then, through the Big M method and the strong duality theorem, the nonlinear terms in the model are linearized, transforming it into a mixed integer linear programming (MILP) model, which is convenient for numerical solution.

[0199] 2. Alternating Iterative Process: Initialize the battery swapping demand baseline curve, the initial prices of the electricity market and frequency regulation market, and generate the initial population of time-of-use battery swapping prices; solve the first-stage battery swapping demand transfer model to obtain the battery swapping demand and backup battery time distribution after the demand transfer; based on backup battery resources, solve the second-stage collaborative bidding and market clearing model to obtain the optimal bidding strategy and market clearing price for the battery swapping station; update the first-stage price input with the new market clearing price and repeat steps S2-S4; when the rate of change of battery swapping service revenue is less than 1%, or the number of iterations reaches the preset upper limit, stop the iteration and output the final bidding strategy.

[0200] According to another embodiment of the present invention, a system for charging and swapping stations involving large-capacity power batteries to participate in the electricity-frequency regulation market bidding is also provided, the system comprising:

[0201] The modeling and load generation module is used to construct an equivalent energy storage aggregation model based on the physical constraints of a large-capacity power battery cluster and a pre-established battery swapping station; the equivalent energy storage aggregation model is used to simulate the probability of electric vehicle battery swapping and generate a baseline curve for disordered battery swapping demand.

[0202] The price guidance and demand optimization module is used to formulate time-of-use battery swapping price signals based on the disordered battery swapping demand baseline curve and combined with the frequency regulation market clearing price, and guide electric vehicle users to adjust their battery swapping time in order to optimize the ideal quantity and time distribution of backup batteries in centralized charging stations.

[0203] The multi-objective optimization solution module is used to establish a minimum dual objective based on the adjusted battery swapping time, constrained by demand fluctuations and profit losses, and uses the particle swarm optimization algorithm to solve for the demand transfer results and backup battery resources.

[0204] The two-layer game bidding module is used to formulate a two-layer game strategy for battery swapping stations to participate in the power and frequency regulation markets based on the demand transfer results and backup battery resources. By introducing multiple constraints, it aims to achieve coordinated allocation of resources and overall revenue in both markets.

[0205] The iterative solution and strategy output module is used to transform the two-level game strategy into a solvable single-level mixed integer linear programming model using optimality conditions and the Big M method, and to output the optimal comprehensive bidding method for the battery swapping station using an alternating iterative algorithm.

[0206] In summary, by utilizing the above-mentioned technical solutions of this invention: 1) A station-network-market physical constraint framework is constructed based on the centralized charging station and distributed distribution station architecture. An equivalent energy storage model is established for large-capacity power batteries, explicitly considering transportation delays, and the disordered battery swapping demand curve is obtained through the Monte Carlo method; 2) An energy and frequency regulation capacity / mileage revenue system is designed, with frequency regulation revenue calculated using both capacity and mileage payments, and time-of-use battery swapping prices are set based on the frequency regulation clearing price to guide demand transfer; 3) A two-layer game of optimal bidding for battery swapping stations and optimal market clearing is established, providing a joint clearing mechanism; 4) A two-stage iterative solution is implemented: the first stage uses the particle swarm optimization (PSO) algorithm to generate time-of-use prices and redundant battery timing sequences, and the second stage uses the KKT theorem and the Big M method to transform the game into a linear model for solution. This invention, by adapting to the characteristics of large-capacity batteries, optimizing the aggregation model and transportation logic, improves frequency regulation capacity and mileage revenue, while reducing unit charging costs and control complexity. An exploratory experiment on the market impact was conducted, and the optimal bidding strategy for battery swapping stations was achieved.

[0207] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for participating in the electricity energy-frequency modulation market bidding by a large-capacity power battery swap station, characterized in that, The method comprises the following steps: Based on the physical constraint framework of large-capacity power battery cluster and pre-established battery swap station, an equivalent energy storage aggregation model is constructed; the equivalent energy storage aggregation model is used to simulate the probability of electric vehicle battery swap, and an unordered battery swap demand benchmark curve is generated; Based on the unordered battery swap demand benchmark curve, combined with the frequency modulation market clearing price, a time-of-use battery swap price signal is formulated, and the electric vehicle users are guided to adjust the battery swap time to optimize the ideal number and time distribution of spare batteries in the centralized charging station; Based on the adjusted battery swap time, a double-objective minimization is established with demand fluctuation and benefit loss as constraints, and a particle swarm algorithm is used to solve the demand transfer result and spare battery resources; According to the demand transfer result and spare battery resources, a double-layer game strategy of battery swap station participating in the electricity market and frequency modulation market is formulated, and through the introduction of multiple constraint conditions, the collaborative allocation and benefit planning of double-market resources are realized; The optimal condition and large M method are used to convert the double-layer game strategy into a solvable single-layer mixed integer linear programming model, and the optimal comprehensive bidding method of the battery swap station is output by using the alternating iteration algorithm.

2. The method of claim 1, wherein, The method comprises the following steps: A physical constraint framework is established based on the double-layer architecture of the battery swap station, wherein the battery swap station comprises a centralized charging station and a distributed distribution station; Through the transportation of the distributed distribution station, the battery and electric vehicle are decoupled in time / space / asset, and the replaced battery is uniformly charged in the centralized charging station and connected to the grid; In the grid-connected process, an equivalent energy storage model is constructed by using the aggregation model and Minkowski addition to aggregate the charging and discharging power and capacity; The probability distribution of the electric vehicle battery swap behavior in each time period in the preset area is simulated by using the Monte Carlo simulation algorithm, and the benchmark value of the battery swap demand in each time period in the unordered state is calculated to generate an unordered battery swap demand benchmark curve.

3. The method of claim 2, wherein, The method comprises the following steps: In the grid-connected process, the aggregation model is used to reduce the difference between the centralized charging station and the replaced battery by using the grid-connected state identification variable, and the overall charging / discharging power of the centralized charging station is aggregated; The grid-connected state variable is introduced to define the grid-connected period / off-grid period of the battery entering / leaving the centralized charging station, and the single battery model domain is expanded to the full scheduling period combined with the charging and discharging characteristics of the large-capacity battery to completely describe the state switching process caused by transportation in the scheduling period; Based on the full scheduling period and the grid-connected state change of the battery, the Minkowski addition is used for processing to obtain the capacity envelope space constraint of the battery cluster, and the influence of the charging / discharging power, the grid-connected period / off-grid period and the grid-connected state change of the battery on the large-capacity battery is considered to obtain the equivalent energy storage model.

4. The method of claim 2, wherein, The Monte Carlo simulation algorithm is used to simulate the probability distribution of the battery swapping behavior of the electric vehicle in each time period of the whole day in the preset area, and the disordered state battery swapping demand benchmark value in each time period is calculated to generate a disordered battery swapping demand benchmark curve. The Monte Carlo simulation algorithm is used to simulate the probability distribution of the battery swapping behavior of the electric vehicle in each time period of the whole day in the preset area, and the disordered state battery swapping demand benchmark value in each time period is calculated to generate a disordered battery swapping demand benchmark curve. The initial battery level, daily driving distance and starting driving time period of each electric vehicle in the preset area are randomly sampled, and the battery consumption dynamics of the electric vehicle during the whole day driving process is simulated. The initial battery level and daily driving distance are determined, and the starting driving time period is sampled. The distance that the vehicle can continue to travel is calculated based on the initial battery level and the power consumption parameters. The actual driving distance from the starting time period is obtained based on the daily driving distance and the cumulative hourly mileage proportion function. When the actual driving distance reaches the distance that the vehicle can continue to travel, it is determined that the electric vehicle will have a battery swapping behavior in the current time period.

5. The method of claim 1, wherein, The disordered battery swapping demand benchmark value in each time period of the whole day is aggregated by counting the battery swapping event time distribution of all electric vehicles, and the probability distribution characteristics of the battery swapping behavior in the preset area are described to generate a disordered battery swapping demand benchmark curve. Based on the disordered battery swapping demand benchmark curve, the time-of-use battery swapping price signal is formulated in combination with the frequency regulation market clearing price, and the electric vehicle users are guided to adjust the battery swapping time to optimize the ideal quantity and time distribution of the spare batteries in the centralized charging station. Based on the difference of the frequency regulation market clearing price, the whole day period is divided into high frequency regulation income period, medium frequency regulation income period and low frequency regulation income period, and differential time-of-use battery swapping prices are formulated accordingly. According to the differential time-of-use battery swapping price signal, the electric vehicle users are guided to shift the battery swapping demand from the high frequency regulation income period to the low income period, and the actual value of the shifted battery swapping demand is obtained to improve the availability of the redundant batteries in the high frequency regulation demand period.

6. The method of claim 1, wherein, The ideal quantity and time distribution of the spare batteries in the centralized charging station are obtained by optimizing the actual value of the shifted battery swapping demand through the sensitivity coefficient of the battery swapping price. Based on the adjusted battery swapping time, a double-objective minimization is established with demand fluctuation and benefit loss as constraints, and a particle swarm algorithm is used to solve the demand transfer result and spare battery resources. The demand fluctuation minimization and equivalent market benefit loss minimization under the time-of-use battery swapping price are taken as double objectives, and a single objective function is integrated by weighted summation through a preset weight coefficient. The particle swarm algorithm is used to solve the single objective function, and the demand transfer result and spare battery resources are output.

7. The method of claim 1, wherein, When the demand transfer result and spare battery resources output by the particle swarm algorithm deviate beyond the tolerance, the price search boundary is automatically scaled and the particle swarm algorithm search process is restarted. Based on the demand transfer result and spare battery resources, a double-layer game strategy of the battery swapping station participating in the electricity market and the frequency regulation market is formulated, and multiple constraint conditions are introduced to realize the collaborative allocation and benefit planning of the double-market resources. The upper layer battery swapping station decision is the bidding decision of the battery swapping station aiming at maximizing the total revenue. The lower layer market clearing is the joint clearing of the power market aiming at minimizing the total purchase cost. The upper layer battery swap station decision and the lower layer market clearing are combined, and by introducing charging and discharging power constraints, equivalent energy storage capacity constraints, frequency modulation resource constraints and bidding price constraints, the collaborative allocation of double market resources and the overall planning of benefits are realized.

8. The method of claim 7, wherein, The multiple constraints include charging and discharging power constraints, equivalent energy storage capacity constraints, frequency modulation resource constraints and bidding price constraints. The charging and discharging power constraints are that the declared charging and discharging power in each period cannot exceed the upper limit of the power corresponding to the equivalent energy storage model, and the charging and discharging states are mutually exclusive. The equivalent energy storage capacity constraints are that the initial power evolution of the equivalent energy storage in each period needs to follow the continuous dynamics considering the charging and discharging efficiency, and needs to be maintained within the upper and lower limits of the capacity. The frequency modulation resource constraints are that the declared frequency modulation capacity is limited by the total available capacity of the aggregated battery, and the declared frequency modulation mileage is associated with the frequency modulation capacity through a mileage coefficient. The bidding price constraints are that the declared energy charging and discharging price, frequency modulation capacity price and mileage price need to be within the preset upper and lower limits of the market price. 9.The method of claim 7, wherein, The optimal comprehensive bidding method of the battery swap station is obtained by using the optimality condition and the big M method to transform the double-layer game strategy into a solvable single-layer mixed integer linear programming model, and using an alternating iteration algorithm. The lower layer market clearing is transformed into equivalent optimality condition constraints by using the optimality condition, and the double-layer game strategy is decomposed into a single-layer optimization problem. The nonlinear complementary relaxation constraints in the single-layer optimization problem are linearized by using the big M method and the strong duality theorem, and a solvable single-layer mixed integer linear programming model is obtained to realize the unified solution of the sequential / joint clearing scenario. The optimal comprehensive bidding method of the battery swap station is obtained by using the alternating iteration algorithm to update the price signal based on the market clearing result, and to backfill the demand transfer result and the standby battery resource.

10. The method of claim 9, wherein the method further comprises: The optimal comprehensive bidding method of the battery swap station is obtained by using the alternating iteration algorithm to update the price signal based on the market clearing result, and to backfill the demand transfer result and the standby battery resource. An unordered battery swap demand baseline curve, an energy and frequency modulation market and a frequency modulation market clearing price are initialized to generate an initial population of time-of-use battery swap prices. The initial population of time-of-use battery swap prices is solved by using a particle swarm optimization algorithm to obtain the battery swap demand after demand transfer and the time distribution of standby batteries. Based on the standby battery resource, a single-layer mixed integer linear programming model is used to solve the optimal bidding strategy of the battery swap station and the market clearing price. When the change rate of the battery swap service benefit is greater than one percent, the price input of the battery swap demand after demand transfer and the time distribution of standby batteries is updated with the new market clearing price, and the particle swarm optimization algorithm and the single-layer mixed integer linear programming model are repeatedly used for solving. When the change rate of the battery swap service benefit is less than one percent, or the iteration number reaches the preset upper limit, the iteration is stopped, and the optimal comprehensive bidding method of the battery swap station is output.

Images