Optimal configuration method of energy storage in wind power collection area considering life cycle and operation strategy

Abstract

The application provides a wind power collection area energy storage optimization configuration method considering cycle life and operation strategy, comprising: obtaining an original scenario based on market electricity price, and eliminating it by using a synchronous back substitution method to obtain a typical operation scenario based on market electricity price, establishing an energy storage cycle life calculation model, obtaining energy storage calendar life and energy storage daily cumulative cycle times, and establishing an energy storage power station operation configuration collaborative optimization double-layer model, wherein the outer model is an energy storage capacity optimization configuration model, and the inner model is an energy storage optimization operation strategy model; and the improved sparrow search algorithm is used to solve the energy storage capacity configuration result considering system frequency modulation demand according to the energy storage power station operation configuration collaborative optimization double-layer model. The wind power collection area energy storage optimization configuration method considering cycle life and operation strategy can reasonably plan energy storage capacity, prolong energy storage service life, improve energy storage operation economy, and help shorten the investment recovery period.

Description

Technical Field

[0001] This invention relates to the field of energy storage power station technology, and in particular to a method for optimizing the configuration of energy storage in wind power collection areas, taking into account cycle life and operation strategy. Background Technology

[0002] Currently, my country's large-scale clustered wind power is developing rapidly. However, the speed of mainstream wind turbine generators is decoupled from the power system frequency, lacking auxiliary frequency regulation capabilities. Their participation in system frequency regulation often comes at the cost of sacrificing maximum power point tracking during regulation periods. By rationally configuring battery energy storage systems, which offer advantages such as high control precision and fast response speed, the primary frequency regulation capability of wind power can be improved, alleviating the wind curtailment problem caused by wind turbine generators participating in frequency regulation. Extensive research has been conducted on the optimized configuration of battery energy storage for system frequency regulation, but current research focuses primarily on energy storage configurations for wind farms, including applications such as compensating for wind power prediction errors and serving as black-start power sources. Regarding primary frequency regulation, because the grid determines the required primary frequency regulation capacity for the next day based on 10% of the predicted maximum power output of the wind farm, the daily required primary frequency regulation capacity for wind farms is constantly changing. Each wind farm's own energy storage configuration inevitably leads to insufficient self-regulation capacity, while other wind farms in the region may have redundant energy storage capacity. Furthermore, the large number of wind farms also presents challenges in scheduling energy storage for individual wind farms. If the spatial smoothing effect of wind power clusters can be fully utilized, configuring energy storage power stations in wind power aggregation areas to meet the primary frequency regulation needs of the power grid will be more conducive to improving the economic efficiency of configuration and the overall frequency regulation capability of the region. Furthermore, energy storage currently faces challenges such as high costs and long investment recovery periods. Various policies have been introduced to allow energy storage to participate in electricity market transactions as a market player. While meeting primary frequency regulation requirements, participating in the electricity market can improve energy storage utilization and shorten its investment recovery period. However, existing research on the optimal configuration of energy storage power stations in wind power aggregation areas, particularly regarding operational strategies, largely fails to consider energy storage lifespan or only provides approximate calculations, resulting in inaccurate and difficult-to-solve results. Therefore, it is essential to design an optimal configuration method for energy storage in wind power aggregation areas that considers cycle lifespan and operational strategies in order to rationally plan energy storage capacity and improve the economic efficiency of energy storage operation. Summary of the Invention

[0003] The purpose of this invention is to provide a method for optimizing the configuration of energy storage in wind power collection areas, taking into account cycle life and operation strategy. This method can rationally plan energy storage capacity, extend the service life of energy storage, improve the economic efficiency of energy storage operation, and help shorten the investment payback period.

[0004] To achieve the above objectives, the present invention provides the following solution:

[0005] A method for optimizing the configuration of energy storage in wind power collection areas, taking into account cycle life and operational strategies, includes the following steps:

[0006] Step 1: Obtain the original scenario based on market electricity prices, and use the synchronous back-substitution elimination method to reduce it, resulting in a typical operating scenario based on market electricity prices, specifically:

[0007] Obtain the original scenarios based on market electricity prices, and for each scenario... Calculate the scenario with the shortest distance using the following formula:

[0008]

[0009] In the formula, τ i For the scene The probability of occurrence, For the scene and scene Based on the Euclidean distance between them, select the scenarios to be deleted according to the following formula.

[0010]

[0011] Modify the remaining number of scenarios, add the probability of the deleted scenario to the nearest scenario, and ensure that the sum of the probabilities of each scenario is 1. Repeat the deletion until the remaining number of scenarios reaches the desired set value H, and complete the reduction to obtain a typical operating scenario based on market electricity price.

[0012] Optionally, in step 2, an energy storage cycle life calculation model is established to obtain the energy storage calendar life and the cumulative number of energy storage cycles per day, specifically:

[0013] Calculate the energy storage calendar lifetime T calendar for:

[0014] T calendar =min{T fl o at ,T cycle} (3)

[0015] In the formula, T float For float charge lifetime, T is a fixed value. cycle To determine the cycle life, an energy storage cycle life calculation model is established to obtain the cycle life T. cycle for:

[0016]

[0017] In the formula, d is the depth of discharge of the energy storage, n(d) is the number of daily cycles of the energy storage at the depth of discharge d, and N is the total energy storage capacity. fail (d) represents the maximum number of cycles the energy storage can perform at the discharge depth d;

[0018] The cumulative daily cycle count of energy storage is calculated, and a mixed-integer linear model for refining the calculation of energy storage cycle life is established. The transition between the charge and discharge states of energy storage in adjacent time periods is represented by G. t Characterization yields:

[0019] G t =|u t -u t-1 | (5)

[0020] In the formula: u t Let u be the charge / discharge state of the stored energy in time period t. t =1 indicates a charging state, u t =0 represents the discharge state. Transforming the inequality constraint into an equivalent linear constraint, we obtain:

[0021] G t =max{0,u t -u t-1}+max{0,u t-1 -u t} (6)

[0022] The state of charge at time t during the energy storage charge / discharge transition is represented as:

[0023]

[0024]

[0025] In equation (7), P t dis P t cha η represents the charging power and discharging power of energy storage in time period t, respectively. c η d These represent the charging and discharging efficiencies of energy storage, respectively. The Big M method is used to process them, introducing auxiliary variables. Add constraint (10):

[0026]

[0027]

[0028] In the formula, M is a large positive number, β∈[0,M], and the discharge depth d of the stored energy in time period t is obtained. t B for:

[0029]

[0030] In the formula, E rateCalculate the conversion factor L for the energy storage cycle count at different depths of discharge to the equivalent cycle count at 100% depth of discharge, based on the rated capacity of the energy storage. eq,t for:

[0031]

[0032] In the formula, The battery energy storage is used for the number of cycles of charging and discharging at 100% depth of discharge until the end of its lifespan, with the depth of discharge d as the decision variable, and the conversion factor L is adjusted accordingly. eq,t After piecewise linearization, we get:

[0033]

[0034] In the formula, K d and B d Piecewise linearization parameters are used for different depths of discharge, and the final calculation of the daily cumulative cycle number N of energy storage is based on the equivalent cycle number statistics. day for:

[0035]

[0036] In the formula, T is the total number of time periods in the scheduling cycle, and L is adjusted according to the depth of discharge. eq,t The parameters are piecewise linearized, and the segment in which the discharge depth of the energy storage system is located can be determined using constraint equation (15):

[0037]

[0038] In the formula, and These are the lower and upper limits of the discharge depth for the e-th segment, respectively. Let the energy storage discharge depth be in the e-th segment during time period t, and introduce auxiliary variables. For located, equal Not located, The value equals 0, where U is the number of segments in the linearization process, thus yielding:

[0039]

[0040] The Big M method was used to process it, and auxiliary variables were introduced. This transforms the mixed-integer nonlinear programming model into a linear one, resulting in:

[0041]

[0042] in,

[0043]

[0044] In the formula, Y is a large positive number, and ω∈[0,Y].

[0045] Step 3: Based on typical operating scenarios based on market electricity prices, energy storage calendar lifespan, and cumulative daily cycle count of energy storage, establish an energy storage capacity optimization configuration model, specifically as follows:

[0046] Based on typical operating scenarios using market-based electricity prices, the calendar lifespan of energy storage, and the cumulative daily cycle count of energy storage, an outer model of a two-layer model for collaborative optimization of energy storage power station operation and configuration is established. The objective function is to maximize the return on energy storage investment over the entire lifespan, and the optimization variable is the configuration capacity of the energy storage system. Specifically:

[0047]

[0048] In the formula, F is the optimal value of the energy storage configuration objective function, and S life I represents the present value of the net income over the entire life cycle of energy storage. year Let be the annual operating profit of the energy storage, D be the number of operating days of the energy storage in a year, f be the optimal value of the expected daily scheduling profit of the energy storage system, E be the configured energy storage capacity, and P be the configured energy storage power. The energy storage capacity investment cost is broken down into an annual depreciation monetary value, C. inv The initial investment cost of energy storage, C om For the annual maintenance cost of energy storage, C the χ is the frequency regulation fee payable to other market participants during the energy storage outage period. E χ represents the unit capacity cost of energy storage. P χ represents the unit power cost of energy storage, r is the discount rate, and χ is the cost per unit of energy storage. omP χ represents the annual unit power operation and maintenance cost of an energy storage system. omE The annual unit capacity operation and maintenance cost of energy storage systems, e the The cost of one frequency regulation per unit capacity, V ob Let EB be the primary frequency regulation capacity that the wind power aggregation area needs to undertake, and EB be the present value annuity factor. Energy storage configuration capacity constraints, energy storage configuration power constraints, and energy storage investment cost constraints are set, with the following conditions:

[0049]

[0050] In the formula, E max P is the maximum capacity configured for energy storage. max C is the maximum power rating configured for energy storage. inv For equipment investment amount, C up This represents the maximum investment amount.

[0051] Step 3: Based on typical operating scenarios based on market electricity prices, energy storage calendar lifespan, and cumulative daily cycle count of energy storage, establish an energy storage optimized operation strategy model, specifically as follows:

[0052] Based on typical operating scenarios using market-based electricity prices, energy storage calendar lifetime, and the cumulative daily cycle count of energy storage, an inner-layer model of a two-layer collaborative optimization model for energy storage power station operation configuration is established. The inner-layer model is an energy storage optimization operation strategy model, with the objective function being to maximize the expected profit of daily scheduling operations. The operating benefits of the energy storage optimization operation strategy model include energy market arbitrage profits, frequency regulation market profits, environmental benefits, government subsidy profits, and primary frequency regulation contract profits. The operating cost is the energy storage system's cycle lifetime decay cost. The specific details of establishing this model are as follows:

[0053]

[0054] In the formula, f is the optimal value of the objective function of the operational strategy model, and π i π j and π q Let I, J, and Q be the probabilities of energy market price scenario i, frequency regulation market capacity price scenario j, and frequency regulation mileage price scenario q, respectively. Let I, J, and Q be the sets of energy market price scenario, frequency regulation market capacity price scenario, and frequency regulation mileage price scenario, respectively. Let h be the operational scenario of energy storage participating in the electricity market. These represent the energy market arbitrage profit and frequency regulation market profit of energy storage in the h-th scenario at time t, respectively. For the environmental benefit in the h-th scenario at time t, For the government subsidy revenue in the h-th scenario during the t-th time period, For the revenue from a frequency modulation contract signed in the h-th scenario, The cost of energy storage system cycle life degradation in the h-th scenario is...

[0055] The arbitrage profits in the energy market are as follows:

[0056]

[0057] In the formula, Let $t$ be the clearing price of the electricity market in the $h$-th scenario and the $t$-th time period. and These represent the discharge and charging power of the energy storage system participating in the energy market during time period t in scenario h;

[0058] FM market revenue includes FM mileage compensation revenue. and AGC capacity compensation revenue Specifically:

[0059]

[0060] In the formula, These represent the frequency modulation mileage price and frequency modulation capacity price for the h-th scenario in the t-th time period, respectively. For the energy storage system's application capacity to participate in the frequency regulation market in the h-th scenario and time period t, γ ave λ represents the average frequency modulation performance index, and λ represents the average frequency modulation mileage.

[0061] The environmental benefits are:

[0062]

[0063] In the formula, The NO required per unit of power generation of thermal power units x SO2 and CO2 emission fees Let be the energy storage discharge power in the h-th scenario at time t.

[0064] The benefits of government subsidies are:

[0065]

[0066] In the formula, e sub The government subsidizes the electricity used for charging energy storage units. Energy storage charging power in the h-th scenario at time t;

[0067] The revenue from a single frequency modulation contract is:

[0068]

[0069] In the formula, e con The contract price per unit capacity for one frequency regulation. This represents the primary frequency modulation capacity required in the h-th scenario during the t-th time period;

[0070] Based on the energy storage lifespan calculation model, the impact of bidding and operation decisions on energy storage lifespan is considered in real time, and the cycle lifespan degradation cost caused by the daily charging and discharging behavior of the energy storage system in the h-th scenario is calculated. Its equivalent power investment cost in terms of loss is:

[0071]

[0072] In the formula, χ P This represents the unit power cost of energy storage;

[0073] Set power balance constraints, power operation constraints, capacity constraints considering self-discharge, capacity management constraints, state of charge / discharge constraints, initial and final state of charge constraints, and cycle life constraints.

[0074] The power balance constraint is as follows:

[0075]

[0076] In the formula, These are the AGC capacity command and the primary frequency modulation capacity command for the h-th scenario at time t, respectively.

[0077] The power operation constraints are:

[0078]

[0079] The capacity constraint considering self-discharge is as follows:

[0080]

[0081] In the formula, σ ess The self-discharge rate of the energy storage system. The available capacity of the energy storage device at time t in the h-th scenario;

[0082] The capacity management constraints are:

[0083]

[0084] In the formula, δ SOC,min and δ SOC,max These are the lower and upper limits of the state of charge of the energy storage system, respectively.

[0085] The charging and discharging state constraints are as follows:

[0086]

[0087] In the formula, These represent the operating status of the energy storage system at time t in the h-th scenario. This indicates that the energy storage system is in a charging state. This indicates that the energy storage system is in a discharging state. and These represent the charging and discharging states of the energy storage system when participating in the energy market;

[0088] The initial and final charge state constraints are as follows:

[0089]

[0090] In the formula, The initial state of charge for the scheduling cycle is given by μ, which is the allowable deviation tolerance of the initial and final states of charge for the cycle.

[0091] The cycle life constraints of the energy storage system are given by equations (5)-(18).

[0092] Step 4: Using the improved sparrow search algorithm, the energy storage capacity configuration considering system frequency regulation requirements is solved based on the two-layer collaborative optimization model of energy storage power station operation configuration. Specifically:

[0093] An adaptive learning factor is added to improve the sparrow search algorithm, where the rate of change z of the fitness value of each individual and the introduced adaptive learning factor are:

[0094]

[0095] In the formula, This represents the position information of the i-th sparrow in the j-th dimension during the t-th iteration. Its fitness value, Let ε be the globally optimal fitness value at the t-th iteration, and let ε be a minimum constant to avoid zero-point error. Let be the adaptive learning factor of the i-th sparrow in the t-th iteration, j = 1, 2, 3, ..., d, where d is the dimension of the variable to be optimized, and z ∈ (0, 2].

[0096] The positions of the discoverer, joiner, and vigilant in the original sparrow search algorithm are updated using the following formula:

[0097]

[0098] In the formula, i termax Let α be the maximum number of iterations, and α be a random number in the range (0,1]. Let L be a 1×d dimensional matrix containing only 1s, and S be a random number that follows a normal distribution. T R1 is the safety value, taken as 0.8; R2 is the warning value, taken as [0,1]. When R2 < S T If R² ≥ S, it indicates that the population is in a safe area, and the discoverer can continue to expand the search area. T This indicates the presence of predators around the population. The person who discovers the predator will immediately sound the alarm and lead all the sparrows to a safe area to forage.

[0099]

[0100] In the formula, X p This is the optimal position for the discoverers so far. Let A be the worst position globally. Let A be a 1×d dimensional matrix where each element is randomly assigned 1 or -1. When i > n / 2, it means that the i-th joiner with a lower fitness value cannot obtain food and needs to fly to other places to find food.

[0101]

[0102] In the formula, ρ and K are step size control parameters, ρ is a random number following a normal distribution with a mean of 0 and a variance of 1, K is a random number taking values ​​in the range [-1, 1], representing the direction of the sparrow's movement, and f i f represents the current fitness value of an individual sparrow. g and f w Let f be the global optimal and worst fitness values, respectively. i >f g This indicates that the sparrow is on the edge of the population and faces the greatest threat. i =f g This indicates that the sparrow is in the middle of the population and has sensed danger, so it will move as close as possible to other sparrows to adjust its search and avoid being attacked;

[0103] The energy storage optimization operation strategy model is solved by using the Gurobi solver on the Yalmip platform and embedding the fitness function of the adaptive sparrow search algorithm to obtain the energy storage capacity configuration result that takes into account the system frequency regulation requirements.

[0104] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects: The wind power aggregation area energy storage optimization configuration method provided by the present invention, which takes into account cycle life and operation strategy, proposes an energy storage collaborative operation strategy. Under the premise of meeting the primary frequency regulation requirements, the operation strategy can be optimized according to the market electricity price, which has good economic efficiency and helps to shorten the investment payback period. The established energy storage cycle life calculation model is embedded into the energy storage operation and configuration collaborative optimization two-layer model, which optimizes the energy storage capacity configuration scheme with the goal of maximizing the return on investment, extends the energy storage life, and thus improves the economic efficiency of the energy storage power station. Attached Figure Description

[0105] 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.

[0106] Figure 1 A schematic diagram illustrating the investment and operation model and departmental structure of an energy storage institution;

[0107] Figure 2 A schematic diagram of a two-layer model architecture for energy storage operation;

[0108] Figure 3 The curve showing the relationship between the number of cycles and the depth of discharge;

[0109] Figure 4 Flowchart for the implementation of the adaptive sparrow search algorithm model;

[0110] Figure 5 A schematic diagram illustrating the charging and discharging power of energy storage for participation in the energy market;

[0111] Figure 6 This diagram illustrates the relationship between the present value of net income over the entire life cycle and energy storage configuration.

[0112] Figure 7 A graph showing the relationship between return on investment and energy storage configuration;

[0113] Figure 8 Power demand diagrams for energy storage participation in frequency regulation in various scenarios;

[0114] Figure 9 To compare the SOC curves of energy storage under different scenarios;

[0115] Figure 10 A graph showing the changes in return on investment under the influence of cost and number of cycles;

[0116] Figure 11 A graph showing the changes in life-cycle revenue under the influence of cost and cycle number factors;

[0117] Figure 12 This is a schematic diagram of the piecewise linearization process at different discharge depths. Detailed Implementation

[0118] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0119] The purpose of this invention is to provide a method for optimizing the configuration of energy storage in wind power collection areas, taking into account cycle life and operation strategy. This method can rationally plan energy storage capacity, extend the service life of energy storage, improve the economic efficiency of energy storage operation, and help shorten the investment payback period.

[0120] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0121] The method described in this invention is for wind power aggregation areas. The energy storage power station in this wind power aggregation area adopts an independent operation mode, and its construction investment can adopt a diversified investment model. The investment and operation model and departmental structure of the energy storage institution are as follows: Figure 1 As shown.

[0122] The wind power cluster energy storage optimization configuration method considering cycle life and operation strategy provided in this embodiment of the invention includes the following steps:

[0123] Step 1: Obtain the original scenario based on market electricity prices, and use the synchronous back-substitution elimination method to reduce it, resulting in a typical operating scenario based on market electricity prices, specifically:

[0124] Energy and ancillary service market prices are crucial factors influencing the optimal allocation of energy storage. While massive amounts of historical price data can effectively describe price fluctuations and magnitudes, the sheer volume of historical scenarios generated based on these prices can lead to excessive computational burden. Therefore, this invention employs a scenario reduction method to decrease the computational burden and complexity of stochastic programming problems. Using one year's worth of price data from the PJM energy market and frequency regulation market as the original scenarios, the historical electricity price scenarios are reduced to H typical scenarios using a synchronous back-substitution elimination method. Specifically:

[0125] Obtain the original scenarios based on market electricity prices, and for each scenario... Calculate the scenario with the shortest distance using the following formula:

[0126]

[0127] In the formula, τ i For the scene The probability of occurrence, For the scene and scene Based on the Euclidean distance between them, select the scenarios to be deleted according to the following formula.

[0128]

[0129] Modify the remaining number of scenarios, add the probability of the deleted scenario to the nearest scenario, and ensure that the sum of the probabilities of each scenario is 1. Repeat the deletion until the remaining number of scenarios reaches the desired set value H, and complete the reduction to obtain a typical operating scenario based on market electricity price.

[0130] Step 2: Establish an energy storage cycle life calculation model to obtain the energy storage calendar life and the cumulative number of energy storage cycles per day, specifically:

[0131] Investment cost and cycle life are two key factors in optimizing energy storage capacity configuration throughout its entire lifecycle. Energy storage calendar life T calendar From float charge life T float and cycle life T cycle The smaller value is determined, that is:

[0132] T calendar =min{T fl o at ,T cycle} (3)

[0133] In the formula, Tfloat Float life refers to the battery's lifespan under normal operating conditions; it is a fixed value, T. cycle Cycle life is related to the charge-discharge behavior of energy storage. The cycle life varies with different depths of discharge (DOD), which in turn affects the investment cost of energy storage throughout its entire lifecycle. Therefore, establishing a cycle life calculation model for energy storage can improve the rationality of energy storage capacity configuration and obtain the cycle life T. cycle for:

[0134]

[0135] In the formula, d is the depth of discharge of the energy storage, n(d) is the number of daily cycles of the energy storage at the depth of discharge d, and N is the total energy storage capacity. fail (d) represents the maximum number of cycles the energy storage can perform at the discharge depth d;

[0136] In engineering, the rainflow counting method is usually used to measure the number of cycles of a battery at different discharge depths. The obtained data is then fitted using methods such as the Nth-order function method, the power function method, and the piecewise fitting method to obtain the corresponding relationship between the two. Taking a certain type of lithium iron phosphate battery as an example, the corresponding data is shown in Table 1. Using the power function method to fit this data, the relationship between the number of battery energy storage cycles N and the discharge depth d can be characterized by equation (5), and the corresponding curve is shown in Table 1. Figure 3 As shown,

[0137] N = 2925d -2.097 +4179 (5)

[0138] Table 1. Relationship between depth of discharge and cycle life

[0139]

[0140] In actual operation, the number of energy storage cycles at different discharge depths needs to be converted to the equivalent number of cycles at 100% discharge depth for statistical analysis. For example, using discharge depth d... t B Perform one complete charge-discharge cycle and calculate the conversion factor L to the equivalent number of cycles at 100% discharge depth for energy storage cycles at different depths of discharge. eq,t for:

[0141]

[0142] In the formula, The battery energy storage is used for the number of cycles of charging and discharging at 100% depth of discharge until the end of its lifespan, with the depth of discharge d as the decision variable, and the conversion factor L is adjusted accordingly. eq,t After piecewise linearization, we get:

[0143]

[0144] In the formula, K d and B d Piecewise linearization parameters for different discharge depths, wherein the specific linearization process is as follows: Figure 12 As shown;

[0145] The cumulative daily cycle count of energy storage is calculated, and a mixed-integer linear model for refining the calculation of energy storage cycle life is established. The transition between the charge and discharge states of energy storage in adjacent time periods is represented by G. t Characterization yields:

[0146] G t =|u t -u t-1 | (8)

[0147] In the formula: u t Let u be the charge / discharge state of the stored energy in time period t. t =1 indicates a charging state, u t =0 represents the discharge state. Transforming the inequality constraint into an equivalent linear constraint, we obtain:

[0148] G t =max{0,u t -u t-1}+max{0,u t-1 -u t} (9)

[0149] The state of charge at time t during the energy storage charge / discharge transition is represented as:

[0150]

[0151]

[0152] In equation (10), P t dis P t cha η represents the charging power and discharging power of energy storage in time period t, respectively. c η d The charging and discharging efficiencies of energy storage are processed using the Big M method, introducing auxiliary variables. Add constraint (13):

[0153]

[0154]

[0155] In the formula, M is a large positive number, β∈[0,M], and the discharge depth of the stored energy in time period t is obtained. for:

[0156]

[0157] In the formula, E rate For the rated capacity of energy storage,

[0158] Combining formulas (6)-(14), the final calculation of the cumulative daily cycle number N of energy storage, using the equivalent cycle number statistics, is obtained. day for:

[0159]

[0160] In the formula, T is the total number of time periods in the scheduling cycle. Since formula (7) has already been applied to L based on the depth of discharge... eq,t The parameters are piecewise linearized, and the segment in which the discharge depth of the energy storage system is located can be determined using constraint equation (15):

[0161]

[0162] In the formula, and These are the lower and upper limits of the discharge depth for the e-th segment, respectively. Let the energy storage discharge depth be in the e-th segment during time period t, and introduce auxiliary variables. For located, equal Not located, The value equals 0, where U is the number of segments in the linearization process, thus yielding:

[0163]

[0164] The Big M method was used to process it, and auxiliary variables were introduced. This transforms the mixed-integer nonlinear programming model into a linear one, resulting in:

[0165]

[0166] in,

[0167]

[0168] In the formula, Y is a large positive number, and ω∈[0,Y].

[0169] like Figure 2As shown, this invention constructs a two-layer model for collaborative optimization of energy storage power station operation and configuration to achieve optimized configuration and operational decisions for energy storage systems. The outer layer is the capacity configuration problem of energy storage power stations considering the uncertainty of market electricity prices, with the objective function being to maximize the return on energy storage investment over the entire life cycle, and the optimization variable being the configured capacity of the energy storage system. The inner layer is the short-term optimized operation problem of the energy storage system considering the cycle life, with the objective of maximizing the expected profit of daily dispatch operation. In the operational decision-making stage, the energy storage system participates in the bidding for the electricity energy market and the frequency regulation ancillary service market. The frequency regulation capacity won in the bid is subject to AGC dispatch instructions, and the energy market capacity operates at the power base point. The energy storage system is considered to be a price taker in the electricity market. The inner and outer layer problems are coupled with each other through interactive optimization variables. On the one hand, the configured capacity of the energy storage system affects the operational decisions of energy storage in the short time scale; on the other hand, the energy storage configuration capacity target in the long time scale needs to be calculated based on the expected profit of daily dispatch in the operational stage. Therefore, overall modeling and decision-making are required, as specifically described in step 3.

[0170] Step 3: Based on typical operating scenarios based on market electricity prices, energy storage calendar lifespan, and cumulative daily cycle count of energy storage, establish an energy storage capacity optimization configuration model, specifically as follows:

[0171] Based on typical operating scenarios using market-based electricity prices, the calendar lifespan of energy storage, and the cumulative daily cycle count of energy storage, an outer model of a two-layer model for collaborative optimization of energy storage power station operation and configuration is established. The objective function is to maximize the return on energy storage investment over the entire lifespan, and the optimization variable is the configuration capacity of the energy storage system. Specifically:

[0172]

[0173] In the formula, F is the optimal value of the energy storage configuration objective function, and S life I represents the present value of the net income over the entire life cycle of energy storage. year Let be the annual operating profit of the energy storage, D be the number of operating days of the energy storage in a year, f be the optimal value of the expected daily scheduling profit of the energy storage system, E be the configured energy storage capacity, and P be the configured energy storage power. The energy storage capacity investment cost is broken down into an annual depreciation monetary value, C. inv The initial investment cost of energy storage, C om For the annual maintenance cost of energy storage, C the χ is the frequency regulation fee payable to other market participants during the energy storage outage period. E χ represents the unit capacity cost of energy storage. P χ represents the unit power cost of energy storage, r is the discount rate, and χ is the cost per unit of energy storage. omP χ represents the annual unit power operation and maintenance cost of an energy storage system. omE The annual unit capacity operation and maintenance cost of energy storage systems, e the The cost of one frequency regulation per unit capacity, Vob Let EB be the primary frequency regulation capacity that the wind power aggregation area needs to undertake, and EB be the present value annuity factor. Energy storage configuration capacity constraints, energy storage configuration power constraints, and energy storage investment cost constraints are set, with the following conditions:

[0174]

[0175] In the formula, E max P is the maximum capacity configured for energy storage. max C is the maximum power rating configured for energy storage. inv For equipment investment amount, C up This represents the maximum investment amount.

[0176] In step 3, based on typical operating scenarios based on market electricity prices, energy storage calendar lifespan, and cumulative daily cycle count of energy storage, an optimized energy storage operation strategy model is established, specifically as follows:

[0177] Based on typical operating scenarios using market-based electricity prices, energy storage calendar lifetime, and the cumulative daily cycle count of energy storage, an inner-layer model of a two-layer collaborative optimization model for energy storage power station operation configuration is established. The inner-layer model is an energy storage optimization operation strategy model, with the objective function being to maximize the expected profit of daily scheduling operations. The operating benefits of the energy storage optimization operation strategy model include energy market arbitrage profits, frequency regulation market profits, environmental benefits, government subsidy profits, and primary frequency regulation contract profits. The operating cost is the energy storage system's cycle lifetime decay cost. The specific details of establishing this model are as follows:

[0178]

[0179] In the formula, f is the optimal value of the objective function of the operational strategy model, and π i π j and π q Let I, J, and Q be the probabilities of energy market price scenario i, frequency regulation market capacity price scenario j, and frequency regulation mileage price scenario q, respectively. Let I, J, and Q be the sets of energy market price scenario, frequency regulation market capacity price scenario, and frequency regulation mileage price scenario, respectively. Let h be the operational scenario of energy storage participating in the electricity market. These represent the energy market arbitrage profit and frequency regulation market profit of energy storage in the h-th scenario at time t, respectively. For the environmental benefit in the h-th scenario at time t, For the government subsidy revenue in the h-th scenario during the t-th time period, For the revenue from a frequency modulation contract signed in the h-th scenario, The cost of energy storage system cycle life degradation in the h-th scenario is...

[0180] The arbitrage profits in the energy market are as follows:

[0181]

[0182] In the formula, Let $t$ be the clearing price of the electricity market in the $h$-th scenario and the $t$-th time period. and These represent the discharge and charging power of the energy storage system participating in the energy market during time period t in scenario h;

[0183] FM market revenue includes FM mileage compensation revenue. and AGC capacity compensation revenue Specifically:

[0184]

[0185] In the formula, These represent the frequency modulation mileage price and frequency modulation capacity price for the h-th scenario in the t-th time period, respectively. For the energy storage system's application capacity to participate in the frequency regulation market in the h-th scenario and time period t, γ ave λ represents the average frequency modulation performance index, and λ represents the average frequency modulation mileage.

[0186] Environmental benefits refer to the benefits of energy storage systems replacing other fossil fuel power generation, indirectly reducing the system's pollution costs, which are:

[0187]

[0188] In the formula, The NO required per unit of power generation of thermal power units x SO2 and CO2 emission fees Let be the energy storage discharge power in the h-th scenario at time t.

[0189] Government subsidies are based on the amount of energy storage charging and discharging, and are as follows:

[0190]

[0191] In the formula, e sub The government subsidizes the electricity used for charging energy storage units. Energy storage charging power in the h-th scenario at time t;

[0192] The revenue from a single frequency modulation contract is:

[0193]

[0194] In the formula, e con The contract price per unit capacity for one frequency regulation. This represents the primary frequency modulation capacity required in the h-th scenario during the t-th time period;

[0195] Based on the energy storage lifespan calculation model constructed above, the impact of bidding and operation decisions on energy storage lifespan is considered in real time, and the cycle lifespan degradation cost caused by the daily charging and discharging behavior of the energy storage system in the h-th scenario is calculated. Its equivalent power investment cost in terms of loss is:

[0196]

[0197] In the formula, χ P This represents the unit power cost of energy storage;

[0198] Set power balance constraints, power operation constraints, capacity constraints considering self-discharge, capacity management constraints, state of charge / discharge constraints, initial and final state of charge constraints, and cycle life constraints.

[0199] The power balance constraint is as follows:

[0200]

[0201] In the formula, These are the AGC capacity command and the primary frequency modulation capacity command for the h-th scenario at time t, respectively.

[0202] The power operation constraints stipulate that the sum of the bidding power of energy storage participating in the electricity market and the primary frequency regulation power must be maintained within the range of the allocated energy storage power:

[0203]

[0204] During battery energy storage operation, there will be a significant energy loss, most of which comes from self-discharge during energy storage. Therefore, the capacity constraint considering self-discharge is as follows:

[0205]

[0206] In the formula, σ ess The self-discharge rate of the energy storage system. The available capacity of the energy storage device at time t in the h-th scenario;

[0207] For energy storage systems to participate in the frequency regulation market, they need to reserve sufficient energy to respond to AGC dispatch commands while meeting primary frequency regulation requirements. The capacity management constraints are as follows:

[0208]

[0209] In the formula, δ SOC,min and δ SOC,max These are the lower and upper limits of the state of charge of the energy storage system, respectively.

[0210] The charging and discharging state constraints are as follows:

[0211]

[0212] In the formula, These represent the operating status of the energy storage system at time t in the h-th scenario. This indicates that the energy storage system is in a charging state. This indicates that the energy storage system is in a discharging state. and These represent the charging and discharging states of the energy storage system when participating in the energy market;

[0213] The initial and final charge state constraints are as follows:

[0214]

[0215] In the formula, The initial state of charge for the scheduling cycle is given by μ, which is the allowable deviation tolerance of the initial and final states of charge for the cycle.

[0216] The cycle life constraints are shown in equations (6)-(19).

[0217] Step 4: Using the improved sparrow search algorithm, the energy storage capacity configuration considering system frequency regulation requirements is solved based on the two-layer collaborative optimization model of energy storage power station operation configuration. Specifically:

[0218] The configuration layer should generate a series of candidate solutions during the optimization process; otherwise, intraday optimization cannot proceed due to lack of operating conditions. Therefore, a heuristic optimization algorithm is needed to meet this requirement. The Sparrow Search Algorithm (SSA) is a novel population optimization algorithm inspired by the foraging and anti-predation behavior of sparrows in the biological world. Compared with other intelligent algorithms such as genetic algorithms commonly used in energy storage configuration problems, it has the characteristics of high accuracy, good stability, and strong robustness. In the Sparrow Search Algorithm, the positions of sparrows in the solution space are randomly distributed. They can move closer to individuals in better positions, but when there are no other individuals nearby, they will adopt a random walk strategy. This strategy will slow down the convergence speed and reduce the convergence accuracy in a finite number of iterations. This invention proposes an improved Sparrow Search Algorithm for model optimization solutions, specifically:

[0219] An adaptive learning factor is added to improve the sparrow search algorithm, where the rate of change z of the fitness value of each individual and the introduced adaptive learning factor are:

[0220]

[0221] In the formula, This represents the position information of the i-th sparrow in the j-th dimension during the t-th iteration. Its fitness value, Let ε be the globally optimal fitness value at the t-th iteration, and let ε be a minimum constant to avoid zero-point error. Let be the adaptive learning factor of the i-th sparrow in the t-th iteration, j = 1, 2, 3, ..., d, where d is the dimension of the variable to be optimized, and z ∈ (0, 2].

[0222] The positions of the discoverer, joiner, and vigilant in the original sparrow search algorithm are updated using the following formula:

[0223]

[0224] In the formula, i termax Let α be the maximum number of iterations, and α be a random number in the range (0,1]. Let L be a 1×d dimensional matrix containing only 1s, and S be a random number that follows a normal distribution. T R1 is the safety value, taken as 0.8; R2 is the warning value, taken as [0,1]. When R2 < S T If R² ≥ S, it indicates that the population is in a safe area, and the discoverer can continue to expand the search area. T This indicates the presence of predators around the population. The person who discovers the predator will immediately sound the alarm and lead all the sparrows to a safe area to forage.

[0225]

[0226] In the formula, X p This is the optimal position for the discoverers so far. Let A be the worst position globally. Let A be a 1×d dimensional matrix where each element is randomly assigned 1 or -1. When i > n / 2, it means that the i-th joiner with a lower fitness value cannot obtain food and needs to fly to other places to find food.

[0227]

[0228] In the formula, ρ and K are step size control parameters, ρ is a random number following a normal distribution with a mean of 0 and a variance of 1, K is a random number taking values ​​in the range [-1, 1], representing the direction of the sparrow's movement, and f i f represents the current fitness value of an individual sparrow. g and f w Let f be the global optimal and worst fitness values, respectively. i >f g This indicates that the sparrow is on the edge of the population and faces the greatest threat. i =f g This indicates that the sparrow is in the middle of the population and has sensed danger, so it will move as close as possible to other sparrows to adjust its search and avoid being attacked;

[0229] The energy storage optimization operation strategy model is solved using the Gurobi solver on the Yalmip platform, embedding an adaptive sparrow search algorithm fitness function to obtain the energy storage capacity configuration result considering system frequency regulation requirements. The specific flowchart is as follows: Figure 4 As shown.

[0230] This invention selects lithium iron phosphate batteries as the electrochemical energy storage type to verify the method described herein. Relevant parameters are shown in Table 2. The number of optimized scenarios is 27, each scenario is 24 hours long, containing 96 time periods, and the USD / CNY exchange rate is 6.8. The energy storage system undertakes primary frequency regulation obligations for a wind farm with a total capacity of 100MW. To simplify the problem, this paper sets the system's primary frequency regulation capacity requirement for the wind farm to be 10% of P. N ;

[0231] Table 2 Simulation-related parameters

[0232]

[0233]

[0234] The inner-layer energy storage operation optimization model was solved using Yalmip and Gurobi algorithms. An energy storage system of 20MW / 60MW·h was selected for operation analysis to verify the effectiveness of the operation strategy. The expected daily revenue was obtained by weighting the revenue from different scenarios according to their respective probabilities. The expected daily net revenue of the energy storage system is 97,629 yuan, including 8,103 yuan from energy market arbitrage, 61,604 yuan from the frequency regulation market, 2,216 yuan from environmental benefits, 12,244 yuan from government subsidies, 24,000 yuan from a primary frequency regulation contract, and 9,831 yuan from cycle life degradation costs. The revenue from the frequency regulation market is the main source of energy storage revenue. Under this operation strategy, the total energy storage revenue far exceeds the cycle life degradation cost; therefore, the energy storage collaborative operation strategy model established in this invention has good economic efficiency.

[0235] The analysis focuses on the scenario with the highest probability, where the charging and discharging power of the energy storage system participating in the energy market is as follows: Figure 5 As shown in the operational strategy model presented in this paper, the energy storage system can not only fulfill the primary frequency regulation obligations of wind farms but also, due to its price sensitivity, utilize its surplus capacity to participate in the energy market and frequency regulation ancillary service market, maximizing profits. The electricity price curve in this scenario shows that the energy storage system can charge when electricity prices are low and discharge when prices are high, thus profiting from "high generation, low storage." Since the energy storage systems participating in frequency regulation ancillary services primarily operate in a shallow charging and discharging state, resulting in lower cycle life decay costs, energy storage systems often choose to participate in the frequency regulation ancillary service market.

[0236] The energy storage power station to be invested and constructed must have the capability of continuous charging time of more than 2 hours, that is, the maximum capacity of the configured energy storage must be more than twice the maximum power. Considering the future development of energy storage technology, the demand for continuous discharge time of energy storage will further increase. Therefore, in the example in this section, P is taken as 0.4E for configuring the energy storage system. The collaborative optimization model of energy storage operation planning is analyzed from the perspectives of net present value of net income over the whole life cycle and return on investment to obtain the optimal configuration target of this project.

[0237] Based on the present value of net income over the entire life cycle, and using the present value of the net income over the entire life cycle of the energy storage system as the allocation target, the relationship between energy storage income and allocation is obtained as follows: Figure 6 As shown, it can be inferred that as the capacity configuration increases from 35MW·h to 70MW·h, the present value of the net income over the entire life cycle of the energy storage system consistently shows an increasing trend, meaning that the larger the energy storage device, the greater the present value of the net profit. This is because the revenue obtained from participating in the frequency regulation auxiliary market far exceeds the cycle life decay cost of energy storage. Furthermore, it does not consider the system's peak-shaving and frequency regulation needs, nor the competitive bidding between the energy storage system and other devices in the region. The energy storage system is only considered as a price taker without regard to the actual clearing results during operation; therefore, the larger the energy storage device, the greater the present value of the net profit. Thus, life-cycle income cannot effectively guide the optimal configuration of energy storage systems.

[0238] Energy storage configuration based on return on investment (ROI): ROI measures the return per unit of investment cost. Using the ROI of energy storage systems as the configuration target, the relationship between ROI and configuration is as follows: Figure 7 As shown, by Figure 7 It can be seen that the project's return on investment (ROI) shows a rapid growth trend in the early stages of energy storage capacity expansion. After reaching its peak, the ROI stabilizes relatively and shows a slight downward trend. Therefore, maximizing the ROI is selected as the energy storage capacity configuration target for this project. In this example, the capacity configuration corresponding to the highest ROI of energy storage is 57.4MW, at which point the ROI is 138.59%. The figure also shows that the energy storage collaborative operation configuration model proposed in this invention has good economic efficiency.

[0239] Capacity configuration analysis considering system energy storage frequency regulation requirements: In actual grid operation, power allocation between energy storage and conventional generating units is usually based on minimizing system operating costs. Due to limitations in energy storage capacity and cost, the system's peak-shaving and frequency regulation pressure still requires the support of flexible resources. Figure 8In a region with a 45% renewable energy penetration rate, the real-time frequency regulation power demand for energy storage in this project is assessed under various scenarios. As shown in the diagram, the system's demand for energy storage scheduling is not constant. Therefore, if energy storage configuration is based solely on electricity price scenarios during actual grid operation, it will overestimate the system's power capacity demand for energy storage and the economic viability of the energy storage project, leading to excessive energy storage equipment and redundancy. However, since the frequency regulation market, which accounts for the largest share of revenue, cannot be participated in at all times, configuring energy storage based solely on maximizing economic efficiency under frequency regulation requirements will inevitably result in energy storage power that cannot meet the system's frequency regulation needs. In this project, the grid company, as the investor in the energy storage system, needs to meet the system's frequency regulation demand for energy storage in long-term planning. Furthermore, with the further increase in renewable energy penetration, the system's demand for frequency regulation power for energy storage is expected to increase. Therefore, the maximum system demand for energy storage power, 11.67MW, is selected as the power configuration for energy storage participation in the electricity market. With the goal of meeting the primary frequency regulation requirements of wind farms and maximizing the return on investment for energy storage projects, an adaptive sparrow search algorithm with nested commercial solvers was used to solve the problem in multiple scenarios. The optimal capacity configuration results are shown in Table 3, Scenario 1.

[0240] Table 3. Energy Storage Optimization Configuration and Operation Results

[0241]

[0242] The following two scenarios are designed for comparison with Scenario 1: Scenario 2, the optimal capacity configuration without considering the energy storage cycle life; Scenario 3: the operational results of the results obtained in Scenario 2 under the strategy in this paper.

[0243] As shown in Table 2, Scenario 2, compared to Scenario 1, the optimal energy storage capacity increases by 4.35MW, but the return on investment decreases by 18.56%. Therefore, energy storage configuration considering cycle life is more reasonable and saves a one-time investment of 6.5249 million yuan. In Scenario 3, compared to Scenario 2, with the same energy storage capacity, the daily expected return decreases by 12,743 yuan, but the cycle life is extended by 272.66 days, and the total life-cycle return on investment increases by 18.18%, demonstrating the effectiveness of the operational strategy considering the cycle life of energy storage proposed in this paper. Scenario 3, compared to Scenario 1, shows that configuring larger energy storage is beneficial for extending the lifespan of energy storage and obtaining greater daily expected returns, but the increased investment cost reduces the return on investment. Therefore, the energy storage collaborative operation configuration model proposed in this paper can balance the lifespan of energy storage and economic efficiency.

[0244] We selected the same scenarios with the highest probabilities in scenarios 2 and 3 for analysis. Figure 9The graphs show the SOC (State of Charge) curves for energy storage under two scenarios. As can be seen from the graphs, the charge-discharge depth of energy storage in scenario 3, which considers cycle life, is shallower than that in scenario 2. This operating mode is more conducive to extending the lifespan of the energy storage. In both scenarios, the SOC of the energy storage returns to its initial value and does not exceed the set upper and lower limits, indicating that it has a good ability to follow frequency modulation signals and can respond well to the frequency modulation requirements of the system in actual operation.

[0245] Under conventional development models, the unit cost of energy storage will decrease with technological advancements, while the number of cycles will increase. Using the energy storage capacity of Scenario 1, this paper analyzes the impact of future changes in energy storage cost and cycle count on the project's return on investment and total lifecycle revenue. The results are as follows: Figure 10 and Figure 11 As shown;

[0246] Depend on Figure 10 It can be seen that as the unit cost of energy storage decreases and the number of cycles increases, the return on investment of the project shows an upward trend, but the change in energy storage cost has a greater impact than the number of cycles.

[0247] Depend on Figure 11 Analysis shows that the impact of changes in energy storage costs on the total lifecycle revenue is approximately constant, while changes in the number of cycles have a more significant impact on revenue in the early stages. This is because as the number of cycles increases, the cost of energy storage degradation during operation decreases, making energy storage systems more willing to participate in market arbitrage, thus resulting in a greater upward trend in revenue. In the later stages, the impact on revenue slows down because the unit cost remains unchanged.

[0248] The present invention provides a wind power cluster area energy storage optimization configuration method that takes into account cycle life and operation strategy. The method proposes an energy storage collaborative operation strategy, which can optimize the operation strategy according to market electricity price while meeting the primary frequency regulation requirements. It has good economic efficiency and helps to shorten the investment payback period. The established energy storage cycle life calculation model is embedded into the two-layer model of energy storage operation and configuration collaborative optimization. The energy storage capacity configuration scheme is optimized with the goal of maximizing the return on investment, extending the energy storage life and thus improving the economic efficiency of the energy storage power station.

[0249] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for optimizing the configuration of energy storage in wind power clusters, taking into account cycle life and operational strategies, characterized in that, Includes the following steps: Step 1: Obtain the original scenario based on market electricity prices, and use the synchronous back-substitution elimination method to reduce it, thereby obtaining the typical operation scenario based on market electricity prices; Step 2: Establish an energy storage cycle life calculation model to obtain the energy storage calendar life and the cumulative number of energy storage cycles per day; Step 3: Based on typical operating scenarios based on market electricity prices, energy storage calendar lifespan, and daily cumulative cycle count of energy storage, establish a two-layer model for collaborative optimization of energy storage power station operation configuration. The outer layer model is an energy storage capacity optimization configuration model with the objective function of maximizing the return on energy storage investment over the entire life cycle and the optimization variable being the energy storage system configuration capacity. The inner layer model is an energy storage optimization operation strategy model with the objective function of maximizing the expected profit of daily scheduling operation. The inner and outer layers are coupled with each other through interactive optimization variables. Step 4: Using the improved sparrow search algorithm, the energy storage capacity configuration result considering the system frequency regulation requirements is obtained by solving the two-layer model of collaborative optimization of energy storage power station operation configuration.

2. The wind power cluster energy storage optimization configuration method considering cycle life and operation strategy according to claim 1, characterized in that, In step 1, the original scenario based on market electricity prices is obtained, and it is reduced using the synchronous back-substitution elimination method to obtain a typical operating scenario based on market electricity prices, specifically: Obtain the original scenarios based on market electricity prices, and for each scenario... Calculate the scenario with the shortest distance using the following formula: （1） In the formula, For the scene The probability of occurrence, For the scene and scene Based on the Euclidean distance between them, select the scenarios to be deleted according to the following formula. : （2） Modify the remaining number of scenarios, add the probability of the deleted scenario to the nearest scenario, and ensure that the sum of the probabilities of each scenario is 1. Repeat the deletion until the remaining number of scenarios reaches the desired set value H, and complete the reduction to obtain a typical operating scenario based on market electricity price.

3. The wind power cluster energy storage optimization configuration method considering cycle life and operation strategy according to claim 2, characterized in that, In step 2, an energy storage cycle life calculation model is established to obtain the energy storage calendar life and the cumulative number of energy storage cycles per day, specifically: Calculate energy storage calendar life for: （3） In the formula, This is the float charge lifetime, a fixed value. To determine the cycle life, an energy storage cycle life calculation model is established to obtain the cycle life. for: （4） In the formula, d represents the depth of discharge of the stored energy. The number of daily cycles for energy storage at the depth of discharge d. The maximum number of cycles for energy storage at the depth of discharge d; The cumulative daily cycle count of energy storage is calculated, and a mixed-integer linear model is established for a refined measurement of energy storage cycle life. This model uses the transition between charge and discharge states of energy storage in adjacent time periods to... Characterization yields: （5） In the formula: This represents the charge / discharge state of the energy stored in time period t. In charging state. In the discharge state, the inequality constraints are transformed into equivalent linear constraints, resulting in: （6） The state of charge at time t during the energy storage charge / discharge transition is represented as: （7） （8） In equation (7), , Let be the charging power and discharging power of the energy storage in time period t, respectively. , These are the charging and discharging efficiencies of energy storage, respectively. The Big M method was used to process it, and auxiliary variables were introduced. Add constraint (10): （9） （10） In the formula, M is a large positive number. The depth of discharge of the stored energy in time period t is obtained. for: （11） In the formula, Calculate the conversion factor for the number of energy storage cycles at different depths of discharge to the equivalent number of cycles at 100% depth of discharge, based on the rated capacity of the energy storage. for: （12） In the formula, The battery energy storage is used for the number of cycles of charging and discharging at 100% depth of discharge until the end of its lifespan, with the depth of discharge d as the decision variable, and the conversion factor is adjusted accordingly. After piecewise linearization, we get: （13） In the formula, and Piecewise linearization parameters were used for different depths of discharge, and the cumulative daily cycle count of energy storage was finally calculated using equivalent cycle count statistics. for: （14） In the formula, T represents the total number of time periods in the scheduling cycle, which is determined based on the depth of discharge. The parameters are piecewise linearized, and the segment in which the discharge depth of the energy storage system is located can be determined using constraint equation (15): （15） In the formula, and These are the lower and upper limits of the discharge depth for the e-th segment, respectively. Let the energy storage discharge depth be in the e-th segment during time period t, and introduce auxiliary variables. , For located, equal , Not located, The value equals 0, where U is the number of segments in the linearization process, thus yielding: （16） The Big M method was used to process it, and auxiliary variables were introduced. This transforms the mixed-integer nonlinear programming model into a linear one, resulting in: （17） in, （18） In the formula, Y is a large positive number. .

4. The wind power cluster energy storage optimization configuration method considering cycle life and operation strategy according to claim 3, characterized in that, In step 3, based on typical operating scenarios based on market electricity prices, energy storage calendar lifespan, and cumulative daily cycle count of energy storage, an energy storage capacity optimization configuration model is established, specifically as follows: Based on typical operating scenarios using market-based electricity prices, the calendar lifespan of energy storage, and the cumulative daily cycle count of energy storage, an outer model of a two-layer model for collaborative optimization of energy storage power station operation and configuration is established. The objective function is to maximize the return on energy storage investment over the entire lifespan, and the optimization variable is the configuration capacity of the energy storage system. Specifically: （19） In the formula, To determine the optimal value of the objective function for energy storage, This represents the present value of the net income over the entire life cycle of energy storage. For the annual operating profit of energy storage, The number of days the energy storage system operates per year. This represents the optimal value for the expected daily operating profit of the energy storage system. For the configured energy storage capacity, For the configured energy storage capacity, The energy storage capacity investment cost is broken down into an annual depreciation monetary value. The initial investment cost of energy storage, For the annual maintenance cost of energy storage, This is a frequency regulation fee that needs to be paid to other market participants during the energy storage outage period. This represents the unit capacity cost of energy storage. This represents the unit power cost of energy storage, where r is the discount rate. The annual unit power operation and maintenance cost of the energy storage system, The annual unit capacity operation and maintenance cost of the energy storage system, The cost per unit capacity for one frequency regulation. This refers to the primary frequency regulation capacity that the wind power aggregation area needs to undertake. The present value factor of an annuity. For float charge life, To determine the energy storage calendar lifetime, set constraints on energy storage configuration capacity, energy storage configuration power, and energy storage investment cost. The constraints are as follows: （20） In the formula, This represents the maximum capacity configured for energy storage. The maximum power rating configured for energy storage. For equipment investment amount, This represents the maximum investment amount.

5. The wind power cluster energy storage optimization configuration method considering cycle life and operation strategy according to claim 4, characterized in that, In step 3, based on typical operating scenarios based on market electricity prices, energy storage calendar lifespan, and cumulative daily cycle count of energy storage, an optimized energy storage operation strategy model is established, specifically as follows: Based on typical operating scenarios using market-based electricity prices, energy storage calendar lifetime, and the cumulative daily cycle count of energy storage, an inner-layer model of a two-layer collaborative optimization model for energy storage power station operation configuration is established. The inner-layer model is an energy storage optimization operation strategy model, with the objective function being to maximize the expected profit of daily scheduling operations. The operating benefits of the energy storage optimization operation strategy model include energy market arbitrage profits, frequency regulation market profits, environmental benefits, government subsidy profits, and primary frequency regulation contract profits. The operating cost is the energy storage system's cycle lifetime decay cost. The specific details of establishing this model are as follows: （21） In the formula, f represents the optimal value of the objective function of the operational strategy model. , and Let I, J, and Q be the probabilities of energy market price scenario i, frequency regulation market capacity price scenario j, and frequency regulation mileage price scenario q, respectively. Let I, J, and Q be the sets of energy market price scenario, frequency regulation market capacity price scenario, and frequency regulation mileage price scenario, respectively. Let h be the operational scenario of energy storage participating in the electricity market. , These represent the energy market arbitrage profit and frequency regulation market profit of energy storage in the h-th scenario at time t, respectively. For the environmental benefit in the h-th scenario at time t, For the government subsidy revenue in the h-th scenario during the t-th time period, For the revenue from a frequency modulation contract signed in the h-th scenario, The cost of energy storage system cycle life degradation in the h-th scenario is... The arbitrage profits in the energy market are as follows: （22） In the formula, Let $t$ be the clearing price of the electricity market in the $h$-th scenario and the $t$-th time period. and These represent the discharge and charging power of the energy storage system participating in the energy market during time period t in scenario h; FM market revenue includes FM mileage compensation revenue. and AGC capacity compensation revenue Specifically: （23） In the formula, , These represent the frequency modulation mileage price and frequency modulation capacity price for the h-th scenario in the t-th time period, respectively. For the energy storage system's application capacity to participate in the frequency regulation market in the h-th scenario and time period t, This refers to the average frequency modulation performance index. Average FM mileage; The environmental benefits are: （24） In the formula, , , These are the costs that each thermal power unit needs to bear for each unit of electricity generated. , , Sewage discharge fees, Let be the energy storage discharge power in the h-th scenario at time t. The benefits of government subsidies are: （25） In the formula, The government subsidizes the electricity used for charging energy storage units. Energy storage charging power in the h-th scenario at time t; The revenue from a single frequency modulation contract is: （26） In the formula, The contract price for a single frequency regulation per unit capacity. This represents the primary frequency modulation capacity required in the h-th scenario during the t-th time period; Based on the energy storage lifespan calculation model, the impact of bidding and operation decisions on energy storage lifespan is considered in real time, and the cycle lifespan degradation cost caused by the daily charging and discharging behavior of the energy storage system in the h-th scenario is calculated. Its equivalent power investment cost is: （27） In the formula, This represents the unit power cost of energy storage; Set power balance constraints, power operation constraints, capacity constraints considering self-discharge, capacity management constraints, state of charge / discharge constraints, initial and final state of charge constraints, and cycle life constraints. The power balance constraint is as follows: （28） In the formula, , These are the AGC capacity command and the primary frequency modulation capacity command for the h-th scenario at time t, respectively. The power operation constraints are: （29） The capacity constraint considering self-discharge is as follows: （30） In the formula, The self-discharge rate of the energy storage system. The available capacity of the energy storage device at time t in the h-th scenario; The capacity management constraints are: （31） In the formula, and These are the lower and upper limits of the state of charge of the energy storage system, respectively. The charging and discharging state constraints are as follows: （32） In the formula, These represent the operating status of the energy storage system at time t in the h-th scenario. This indicates that the energy storage system is in a charging state. This indicates that the energy storage system is in a discharging state. and These represent the charging and discharging states of the energy storage system when participating in the energy market; The initial and final charge state constraints are as follows: （33） In the formula, The initial state of charge for the scheduling cycle. The allowable deviation tolerance for the state of charge at the beginning and end of the cycle; The cycle life constraints of the energy storage system are given by equations (5)-(18).

6. The wind power cluster energy storage optimization configuration method considering cycle life and operation strategy according to claim 5, characterized in that, In step 4, an improved sparrow search algorithm is used to solve for the energy storage capacity configuration considering system frequency regulation requirements based on the two-layer model of collaborative optimization of energy storage power station operation configuration. Specifically: An adaptive learning factor is added to improve the sparrow search algorithm, where the rate of change z of the fitness value of each individual and the introduced adaptive learning factor are: （34） In the formula, This represents the position information of the i-th sparrow in the j-th dimension during the t-th iteration. Its fitness value, Let be the globally optimal fitness value at the t-th iteration. To minimize constant error and avoid zero-point error, Let be the adaptive learning factor for the i-th sparrow in the t-th iteration. d is the dimension of the variable in the problem to be optimized. ; The positions of the discoverer, joiner, and vigilant in the original sparrow search algorithm are updated using the following formula: （35） In the formula, The maximum number of iterations, A random number in the range (0,1]. Let L be a random number that follows a normal distribution. S is a matrix whose all elements are 1. T R1 is a safety value, set to 0.8; R2 is a warning value, set to [0,1]. If this indicates the population is in a safe area, the discoverer can continue to expand the search area. This indicates the presence of predators around the population. The person who discovers the predator will immediately sound the alarm and lead all the sparrows to a safe area to forage. （36） In the formula, This is the optimal position for the discoverers so far. Let A be the worst position globally. A dimensional matrix, where each element is randomly assigned 1 or -1, when When the i-th participant with the lowest fitness value is unable to obtain food, it needs to fly to other places to forage. （37） In the formula, K is the step size control parameter. It is a random number that follows a normal distribution with a mean of 0 and a variance of 1. K is a random number taking the value [-1, 1], representing the direction in which the sparrow moves. This represents the current fitness value of an individual sparrow. and These are the globally optimal and worst fitness values, respectively. This indicates that the sparrow is on the edge of the population and is under the greatest threat. This indicates that the sparrow is in the middle of the population and has sensed danger, so it will move as close as possible to other sparrows to adjust its search and avoid being attacked; The energy storage optimization operation strategy model is solved by using the Gurobi solver on the Yalmip platform and embedding the fitness function of the adaptive sparrow search algorithm to obtain the energy storage capacity configuration result that takes into account the system frequency regulation requirements.

Images