Network construction type energy storage scheduling method considering mechanism characteristics of lithium battery
By considering the mechanism characteristics of lithium batteries, obtaining aging and decay and economic value coefficients, and optimizing the network-type service of lithium battery energy storage, the problem that the dynamic characteristics of lithium battery energy storage in the existing technology is difficult to reflect, and the scheduling efficiency of lithium battery energy storage is improved economic benefits of network-type services.
Patent Information
- Application Number
- CN202411010266.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-26
- Publication Date
- 2024-11-12
AI Technical Summary
The prior art is difficult to accurately reflect the real dynamic characteristics of lithium battery energy storage in network-type scenarios, resulting in insufficient physical feasibility and operational efficiency of the scheduling scheme.
By obtaining the mapping relationship between the aging decline of lithium battery energy storage and the state quantity on the dispatch side, determine the economic value coefficient and aging cost function, optimize the energy cost and benefit function when lithium battery energy storage provides network-type services, and combine the uncertainty of the grid frequency. , Benders algorithm is used to solve it to maximize economic benefits and obtain the dispatching instructions for network-type lithium battery energy storage.
The physical feasibility and operation efficiency of the lithium battery energy storage scheduling scheme are improved, and the proportion and power of lithium battery energy storage provide network-type services on a long-term scale is optimized, so as to achieve effective scheduling and control of network-type energy storage.
Smart Images

Figure CN118944158A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of energy storage scheduling, and in particular to a grid-type energy storage scheduling method that takes into account the mechanism characteristics of lithium batteries. Background Art
[0002] At present, how to use the rapid response capability of lithium battery energy storage systems to provide the dynamic performance required for grid stability has attracted much attention. In view of this, grid-connected energy storage provides an effective solution based on the simulation of the external characteristics of the dynamic performance output of thermal power units.
[0003] Existing technologies mainly study how power electronic devices in grid-type energy storage are physically implemented from a control perspective on a short-term scale. However, on a longer time scale such as day-ahead dispatch, few technologies regulate the economic and safe operation of grid-type energy storage from the perspective of energy storage operation.
[0004] In addition, as a source-side device, lithium battery energy storage itself has multiple types of dynamic characteristics in the process of providing grid-building services, and it is necessary to use an effective model to model and describe its operating dynamic characteristics over a longer time scale. However, in the research on grid-building energy storage, most of the modeling of lithium battery energy storage itself adopts simplified or equivalent processing, lacking consideration of the mechanism characteristics of lithium battery energy storage itself in the grid-building scenario, and it is difficult to accurately reflect the real grid-building dynamic characteristics of lithium battery energy storage. Summary of the invention
[0005] The present application aims to solve one of the technical problems in the related art at least to some extent.
[0006] To this end, the purpose of this application is to propose a grid-type energy storage scheduling method that takes into account the mechanism characteristics of lithium batteries, which solves the technical problem that the existing methods are difficult to accurately reflect the real dynamic characteristics of lithium battery energy storage. By considering the mechanism characteristics of lithium batteries, the physical feasibility of the scheduling scheme and the operating efficiency of lithium battery energy storage are effectively improved.
[0007] To achieve the above-mentioned purpose, the embodiment of the present application proposes a grid-forming energy storage scheduling method that takes into account the mechanism characteristics of lithium batteries, including: obtaining a mapping relationship between the aging and decay amount of lithium energy storage and the state amount of the scheduling side; determining the economic value coefficient of the grid-forming service provided by the lithium energy storage; determining the aging cost function when the lithium energy storage provides grid-forming services based on the mapping relationship, determining the grid-forming service revenue function when the lithium energy storage provides grid-forming services based on the economic value coefficient, and determining the energy cost function and constraints when the lithium energy storage provides grid-forming services; determining the economic benefit function of the lithium energy storage under the uncertainty of the grid frequency based on the aging cost function, the grid-forming service revenue function and the energy cost function, considering the constraints, solving the problem with the goal of maximizing the economic benefit of the lithium energy storage, and obtaining a grid-forming coefficient scheduling instruction set for the grid-forming lithium energy storage.
[0008] The grid-connected energy storage scheduling method that takes into account the mechanism characteristics of lithium batteries in the embodiment of the present application optimizes the proportion and power of lithium energy storage to provide multiple types of grid-connected services on a day-ahead scale in view of the uncertainty of grid frequency and the dynamic characteristics of grid-connected services, thereby realizing scheduling control of grid-connected energy storage. By considering the mechanism characteristics of lithium batteries in the optimal scheduling model, the mechanism-driven aging characteristics and battery constraints are incorporated into the optimization model to provide a more efficient lithium energy storage scheduling solution.
[0009] Optionally, in one embodiment of the present application, obtaining a mapping relationship between a lithium battery energy storage aging decay amount and a scheduling side state amount includes:
[0010] Step S1: obtaining an electrochemical model of lithium battery energy storage;
[0011] Step S2: obtaining the state of charge level and constant current excitation through uniformly distributed sampling within the feasible domain;
[0012] Step S3: for the electrochemical model, given the state of charge level, applying constant current excitation within a preset decision step to achieve simulation of the electrochemical model;
[0013] Step S4: when simulating the electrochemical model, the side reaction intensity at each sampling moment is obtained at a sampling step of seconds, and the aging decay amount related to the side reaction is calculated based on the side reaction intensity obtained by sampling according to the parameters related to the electrode material in the electrochemical model, wherein the side reaction intensity includes the solid electrolyte interface film growth rate, the lithium precipitation rate and the active material loss rate;
[0014] Step S5: accumulating the aging decay amount related to the side reaction over the complete simulation length to obtain the aging decay amount on the decision step scale;
[0015] Step S6: when simulating the electrochemical model, the output voltage at each sampling moment is obtained at a sampling step of seconds, and the charge and discharge power at the decision step scale is determined based on the constant current excitation and the output voltage;
[0016] Step S7: repeating steps S2-S6 to obtain multiple sets of simulation data, wherein each set of simulation data includes a state of charge level, a charge and discharge power, and an aging degradation amount;
[0017] Step S8: With the state of charge level and the charge and discharge power as independent variables and the aging degradation amount as the dependent variable, a fitting formula is obtained by using the piecewise linearized convex hull to perform fitting based on multiple sets of simulation data.
[0018] Optionally, in one embodiment of the present application, the aging decay amount associated with the side reaction is:
[0019] τ SEI ∝j SEI ,τ LP ∝j LP ,τ AM ∝j AM
[0020] Among them, τ SEI is the aging decay caused by the growth of the solid electrolyte interface film, j SEI is the solid electrolyte interface film growth rate, τ LP is the aging decay caused by lithium deposition, j LP is the lithium deposition rate, τ AM is the aging decay due to loss of active materials, j AM is the active material loss rate;
[0021] The aging decay amount on the decision step size scale is:
[0022]
[0023] Among them, N t is the preset decision step size;
[0024] If the constant current excitation amplitude I0 ≥ 0, the charging power P on the decision step scale B,c is 0, the discharge power P B,d P L , if I0<0, the charging power P on the decision step scale B,c P L , discharge power P B,d is 0, where
[0025]
[0026] Among them, V tis the output voltage at sampling time t;
[0027] The fitting formula is expressed as:
[0028]
[0029] in, is the mth segment, expressed as:
[0030]
[0031] Among them, C S is the state of charge, is the coefficient obtained by plane fitting of the mth segment.
[0032] Optionally, in one embodiment of the present application, determining the economic value coefficient of the grid-building service provided by the lithium battery energy storage includes:
[0033] Obtain historical external power grid frequency data for lithium battery energy storage to provide grid-building services;
[0034] Based on the historical external power grid frequency data at each moment in the preset time period at the second-level sampling step, determine the external power grid frequency deviation of the primary frequency regulation service, damping and inertia response at each moment in the preset time period;
[0035] According to the frequency deviation of the external power grid within the preset time period, the economic value brought to the power grid when the lithium battery energy storage provides 1 unit of primary frequency regulation service, 1 unit of damping, and 1 unit of inertia is determined respectively, and the determined 1 unit of economic value is used as the economic value coefficient of the corresponding grid-building service.
[0036] Optionally, in one embodiment of the present application, the external power grid frequency deviations of the primary frequency regulation service, damping and inertia response at each moment in the preset time period are respectively:
[0037]
[0038] in, is the external grid frequency deviation of the primary frequency regulation service response at time t, is the external grid frequency deviation of the damping response at time t, is the external grid frequency deviation of the inertia response at time t, ω t is the external power grid frequency data at each moment, ω0 is the rated value of the power grid frequency, ω s is the external grid frequency deviation of the primary frequency modulation response after equivalent low-pass filtering, expressed as:
[0039]
[0040] Among them, Δωt-i is the frequency difference ω outside the primary frequency modulation service response dead zone at time ti t -ω0, t s is the response time constant of the primary frequency modulation service, t c is the cutoff constant;
[0041] The economic value of one unit of primary frequency regulation service is:
[0042]
[0043] Among them, | αr is 1-α of the frequency deviation distribution r Confidence interval, is the average on-grid electricity price, N T is the total number of decision points;
[0044] The economic value of 1 unit damping is:
[0045]
[0046] The economic value of 1 unit of inertia is:
[0047]
[0048] Among them, r I,init The economic value associated with inertia starting is given by the slope obtained by fitting a linear function using the starting fuel cost calculated with the hot cavity starting of different traditional thermal power units as the independent variable and the inertia coefficient of the thermal power unit as the dependent variable.
[0049] Optionally, in one embodiment of the present application, the aging cost function is:
[0050]
[0051] Among them, r Deg is the aging price of battery cell per unit capacity, N c is the number of cells in the lithium battery energy storage, It is given by the fitting formula according to the state of charge and power at the Tth moment;
[0052] The revenue function of networking services is expressed as:
[0053] z GFS =r I κ I +r D κ D +r PFR κ PFR
[0054] Among them, κ I , κD , κ PFR The coefficient that provides inertia, damping, and primary frequency regulation services for lithium battery energy storage, κ I , κ D , κ PFR Construct a network-based scheduling instruction set κ, κ = {κ PFR ,κ D ,κ I};
[0055] The energy cost function is expressed as:
[0056]
[0057] in, is the on-grid electricity price at time T, is the electricity purchase price at time T, It is the peak-shaving charging and discharging power of lithium battery energy storage at each decision point T in a day.
[0058] Optionally, in one embodiment of the present application, the constraint condition includes a state of charge constraint, a power opportunity constraint and a power constraint, and the power opportunity constraint is expressed as:
[0059]
[0060] Among them, Pr{·} represents the probability, N t is the preset decision step size;
[0061] according to The linear characteristics of the power opportunity constraint are converted into a power constraint, and the power constraint is expressed as:
[0062]
[0063] in, is the grid-connected power of lithium battery energy storage within the Tth decision step, is the real-time grid-connected power of lithium battery energy storage within the Tth decision step α that constitutes the probability distribution ω Quantile, α ω is the confidence level, η d is the discharge efficiency of lithium battery energy storage, η c The efficiency of the charging side of lithium battery energy storage, is the state of charge at the current decision point Determine the feasible domain of charging and discharging power output, and the real-time grid-connected power of lithium battery energy storage It is expressed as:
[0064]
[0065] in, The frequency response power of lithium battery storage participating in the grid-building service at time t, is the discharge power of lithium battery energy storage participating in peak regulation at time t, is the charging power of lithium battery energy storage participating in peak load regulation at time t;
[0066] The state of charge constraint is expressed as:
[0067]
[0068] in, It is the lower limit of the lithium battery energy storage charge state. is the lithium battery energy storage charge state at the Tth decision step, is the upper limit of the lithium battery energy storage charge state, and the lithium battery energy storage charge state sequence vector C S for:
[0069]
[0070] Among them, P b,d is the discharge power sequence vector, P b,c is the charging power sequence vector, P b,c for P b,d for
[0071]
[0072] Among them, the power of lithium battery energy storage grid-connected discharge for Part of the power of lithium battery energy storage grid charging for part, It is the internal output discharge power of lithium battery energy storage. Output charging power for the internal lithium battery energy storage.
[0073] Optionally, in one embodiment of the present application, the economic benefit function of lithium battery energy storage under grid frequency uncertainty is:
[0074]
[0075] Where Ω is the uncertainty set of the power grid frequency data, Ω={ω1,ω2,…,ω n}, Represents a data sequence vector of the power grid frequency for one day, P G is the peak load power sequence vector,
[0076] Optionally, in one embodiment of the present application, the process of using the Benders algorithm to solve and obtain the meshing coefficient scheduling instruction includes:
[0077] Step S1: split the grid-connected lithium battery energy storage scheduling model into a main problem MP considering economic benefits and multiple sub-problems SP considering aging costs, energy costs and constraints corresponding to multiple scenarios under uncertain grid frequencies;
[0078] Step S2: Solve the initial MP to obtain the optimal solution of MP;
[0079] Step S3: Based on the optimal solution of MP, determine all corresponding ω i The SP problem under ∈Ω, and the optimal solution of MP is brought in as the determined value, and the constraints are transformed into the standard form of constraints;
[0080] Step S4: Solve for all ω i When solving the SP problem under ∈Ω, if the current SP problem is not feasible, obtain the corresponding polar ray Add polar rays to Benders cut set The corresponding feasible cut, if all SP problems have optimal solutions, is based on the average value y of the optimal solutions of each SP problem * Update the upper bound UB and obtain all ω i The extreme points of SP problem under ∈Ω Adding vertices to the Benders cut set The corresponding optimal cut;
[0081] Step S5: Solve the MP containing the Benders cut set, obtain the optimal solution of the MP, and update the lower bound LB to the optimal value of the MP containing the Benders cut set;
[0082] Step S6: Repeat steps S3-S5 until |UB-LB|≤∈, and take the optimal solution of the MP problem containing the Benders cut set as the optimal solution of the scheduling problem, and use it as the scheduling instruction of the network configuration coefficient. s , bring in the grid-forming coefficient scheduling instruction as the determined value, solve the SP corresponding to the scenario to obtain the optimal solution, and obtain the charging and discharging power of the lithium battery energy storage at each moment to form a charging and discharging power operation curve.
[0083] Optionally, in one embodiment of the present application, the main problem MP is expressed as:
[0084]
[0085] 0≤κ≤κ max
[0086] in, For all uncertain scenarios α that constitutes the probability distribution ω Quantile, κ max It is the maximum value of the network scheduling instruction;
[0087] For ω i The subproblem SP under ∈Ω is expressed as:
[0088]
[0089] st power constraint
[0090] State of Charge Constraints
[0091] Among them, κ * is the optimal solution of MP;
[0092] The constraints in standard form are expressed as:
[0093]
[0094] Among them, I is the unit variable, H K , H G is the partial multiplier of the constraint matrix, and h is the rearranged right-hand side term;
[0095] If the SP problem currently being solved is infeasible, the polar ray added to the Benders cut set Γ The corresponding feasible cut is:
[0096]
[0097] Among them, h i is the right-hand side term of the constraint condition in the i-th scenario, is the partial multiplier of the constraint matrix in the i-th scenario;
[0098] The upper bound UB is updated to:
[0099] UB=-z GFS (κ * )+y *
[0100] If all SP problems have optimal solutions, the poles added to the Benders cut set Γ The corresponding optimal cut is:
[0101]
[0102] Among them, θ is an auxiliary variable, γ i is the probability of the i-th scenario in Ω;
[0103] The MP containing the Benders cut set is:
[0104]
[0105] 0≤κ≤κ max
[0106] Benders cut set Γ
[0107] The lower bound LB is updated to:
[0108] LB=-z GFS (κ * )+θ *
[0109] Among them, θ * The optimal auxiliary variables obtained for solving MP with Benders cut sets.
[0110] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0111] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:
[0112] Figure 1 A schematic flow chart of a grid-type energy storage scheduling method taking into account the mechanism characteristics of lithium batteries provided in Example 1 of the present application;
[0113] Figure 2 A flow chart of another grid-type energy storage scheduling method taking into account the mechanism characteristics of lithium batteries provided in an embodiment of the present application. DETAILED DESCRIPTION
[0114] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.
[0115] The following describes a grid-type energy storage scheduling method that takes into account the mechanism characteristics of lithium batteries according to an embodiment of the present application with reference to the accompanying drawings.
[0116] Figure 1 A flow chart of a grid-type energy storage scheduling method taking into account the mechanism characteristics of lithium batteries provided in Example 1 of the present application.
[0117] like Figure 1As shown, the grid-type energy storage scheduling method considering the mechanism characteristics of lithium batteries includes the following steps:
[0118] Step 101, obtaining a mapping relationship between the aging and decay amount of lithium battery energy storage and the state amount of the dispatching side;
[0119] Step 102, determining the economic value coefficient of the grid-building service provided by lithium battery energy storage;
[0120] Step 103, determining the aging cost function of the lithium battery energy storage when providing the networking service based on the mapping relationship, determining the networking service revenue function of the lithium battery energy storage when providing the networking service based on the economic value coefficient, and determining the energy cost function and constraint conditions of the lithium battery energy storage when providing the networking service;
[0121] Step 104, based on the aging cost function, the grid-forming service revenue function and the energy cost function, the economic benefit function of the lithium battery energy storage including the grid frequency uncertainty is determined, and the constraints are considered, and the economic benefit maximization of the lithium battery energy storage is taken as the goal to obtain the grid-forming coefficient scheduling instruction set of the grid-forming lithium battery energy storage.
[0122] The grid-connected energy storage scheduling method that takes into account the mechanism characteristics of lithium batteries in the embodiment of the present application optimizes the proportion and power of lithium energy storage to provide multiple types of grid-connected services on a day-ahead scale in view of the uncertainty of grid frequency and the dynamic characteristics of grid-connected services, thereby realizing scheduling control of grid-connected energy storage. By considering the mechanism characteristics of lithium batteries in the optimal scheduling model, the mechanism-driven aging characteristics and battery constraints are incorporated into the optimization model to provide a more efficient lithium energy storage scheduling solution.
[0123] See also Figure 2 , Figure 2 A flow chart of another grid-type energy storage scheduling method considering the mechanism characteristics of lithium batteries provided in an embodiment of the present application is shown. The method is executed by an electronic device.
[0124] The grid-type energy storage scheduling method considering the mechanism characteristics of lithium batteries of the present invention specifically includes: defining the discharge of lithium battery energy storage as positive power and charging as negative power, defining the decision step length as ΔT, the time label value of the decision point as T, and the total number of decision points as N T , the second-level sampling step is Δt, the second-level data point time label value is t, and the total number of second-level sampling points contained in a single decision step is N t .
[0125] like Figure 2 As shown, the method comprises the following steps:
[0126] 1) Based on the simulation of the lithium battery electrochemical model, the mapping relationship between the aging and decay of the lithium battery energy storage and the state quantity of the dispatching side is obtained, and it is piecewise linearized into an aging characteristic function with the lithium battery energy storage state of charge and the charging and discharging power as independent variables;
[0127] 2) Based on the dynamic characteristic equation of lithium battery energy storage providing grid-building services and historical grid frequency data, the economic value coefficient of lithium battery energy storage providing grid-building services is obtained;
[0128] 3) Obtain the grid-building service revenue function, energy cost function, and aging cost function when lithium battery energy storage provides grid-building services, and obtain the state of charge constraints including reaction mechanisms when lithium battery energy storage provides grid-building services and the power constraints during the interaction with the external power grid;
[0129] 4) Solve the stochastic optimization problem of grid-connected lithium battery energy storage scheduling and operation with grid frequency uncertainty, and give the grid-connected coefficient scheduling instructions and power operation curve of grid-connected lithium battery energy storage.
[0130] In some embodiments, step 1) specifically includes:
[0131] The lithium battery electrochemical model can be used to simulate the internal and external states of lithium batteries with deterministic physical entities. In the electrochemical model, the set parameters corresponding to the deterministic physical entity of the network-type lithium battery energy storage include electrode materials and ambient temperature. For the electrochemical model of the determined physical entity, the electrochemical model can be used to analyze the main side reaction intensity inside it, including: solid electrolyte interface film growth rate j SEI , lithium deposition rate j LP , active material loss rate j AM .
[0132] For an electrochemical model with determined physical entities, at a given state of charge level C S , within the simulation length of a decision step, a constant current excitation is applied to the lithium battery energy storage By simulating the electrochemical model at a sampling step of seconds, the side reaction intensity level j at each sampling time t can be obtained. t,SEI ,j t,LP ,j t,AM Furthermore, according to the parameters related to the electrode material in the electrochemical model, the aging decay amount related to the side reaction can be calculated from the side reaction intensity, which is a positive proportional relationship:
[0133] τ SEI ∝j SEI ,τ LP ∝j LP ,τ AM ∝j AM
[0134] In the formula, τSEI , τ LP , τ AM Respectively represent the aging decay caused by various types of side reactions, and its unit dimension is ampere-hour. Similarly, the aging decay at each time point can be obtained in the time series simulation of the electrochemical model at the second-level sampling point, and is represented by the superscript r. The aging decay is accumulated over the entire simulation length to obtain the cumulative aging decay at the decision step scale:
[0135]
[0136] At the same time, the output voltage V at each sampling time t can also be obtained through the electrochemical model t Based on the applied constant current excitation, the power at the decision step size can be obtained, corresponding to the average power over the full simulation length:
[0137]
[0138] Within the feasible region, a new state of charge level C is obtained by uniformly distributing sampling. S and current excitation amplitude I0, adjust C S Repeat the above steps after I0 to obtain the new cumulative aging decay and average power on the decision step scale, and record C at this time S , I0 and τ L , P L , and the power is divided into charging and discharging, that is, if I0≥0, the charging power P B,c =0, discharge power P B,d =P L , otherwise if I0<0, the charging power P B,c =P L , discharge power P B,d = 0. When more than 200 groups of C have been uniformly obtained within the feasible region S , I0 data points, there are more than 100 sets of data points I0 ≥ 0, more than 100 sets of data points I0 < 0, then use C S , P B,c , P B,d As an independent variable, τ L As the dependent variable, the piecewise linearized convex hull is used for fitting, and the fitting formula is as follows:
[0139]
[0140] In the formula represents the mth segment, which represents a fitting plane, and the formula is as follows:
[0141]
[0142] In the formula is the coefficient obtained by plane fitting of the mth segment. The above formula is the aging characteristic function obtained by taking the state of charge of lithium battery storage and the charge and discharge power as independent variables. At this time, on the decision step scale, when a certain state of charge level C is given S And lithium battery energy storage charging and discharging power P B,c , P B,d When , the aging decay amount can be directly obtained by the piecewise fitting formula.
[0143] In some embodiments, step 2) specifically includes:
[0144] The grid-connected lithium battery energy storage provides primary frequency regulation service, damping and inertia according to the frequency deviation of the external power grid. The coefficients indicating the amount of service provided are expressed as κ PFR , κ D , κ I The rated value of the grid frequency is ω0. At the sampling step of seconds, the external grid frequency data at each moment is expressed as ω t The frequency deviation of primary frequency regulation service, damping and inertia response is expressed as Δω PFR , Δω D , Δω I .
[0145] Based on the external power grid frequency data at each moment, the above frequency deviation is obtained, and the frequency deviation value needs to be scaled to a similar numerical range, as shown in the following formula:
[0146]
[0147] In the formula Represents the time derivative of frequency data, which can be approximated by discrete method: ω s It is the frequency deviation of the primary frequency modulation response after the equivalent low-pass filtering, as shown in the following formula:
[0148]
[0149] Where Δω t-i is the frequency difference ω outside the dead zone of the primary frequency modulation response at the tith moment t -∈0, where t s is the response time constant of primary frequency modulation. Further, let t c is the cutoff constant, i>t c The higher-order terms can be ignored, so ω s It can be obtained by summing up the finite terms:
[0150]
[0151] Based on historical grid frequency data, the frequency deviation of various types of grid-building service responses over a period of time can be calculated. Considering the substitutability of grid-building lithium battery storage for traditional thermal power in terms of primary frequency regulation, damping and inertia, traditional thermal power has a more stable power generation output space without reserving a portion of power generation capacity for providing the above-mentioned grid-building services. Therefore, the economic value of various types of grid-building services can be equivalent to the opportunity cost of traditional thermal power when providing the same amount of grid-building services. According to the stability of the frequency deviation distribution, the frequency deviation of one month in the historical data is taken to form a probability distribution, and let | αr 1-α represents the frequency deviation distribution r Confidence interval, let is the average on-grid electricity price, then the economic value of primary frequency regulation service is:
[0152]
[0153] The formula implies a frequency modulation unit coefficient κ PFR =1, r PFR It represents the economic value of lithium battery energy storage to the power grid when it provides 1 unit of primary frequency modulation. Similarly, it can be obtained that the economic value of lithium battery energy storage to the power grid when it provides 1 unit of damping is r D :
[0154]
[0155] However, in traditional thermal power, inertia essentially represents mechanical properties related to kinetic energy. Therefore, the startup process of traditional thermal power, that is, the starting process of the rotor, needs to be considered in the economic value of inertia. The startup fuel cost of different traditional thermal power units calculated by starting with the hot cavity is obtained from the data. The startup fuel cost is used as the independent variable and the inertia coefficient of the thermal power unit is used as the dependent variable for linear function fitting. The slope is the economic value r related to inertia startup. I,init , then the economic value r that lithium battery energy storage brings to the power grid when providing 1 unit of inertia I for:
[0156]
[0157] In some embodiments, step 3) specifically includes:
[0158] The cost-benefit structure of network-based lithium batteries includes: the benefits of network-based services GFS , grid-connected energy cost z G , aging cost z Deg .z GFS Economic value from different types of networking services:
[0159] z GFS =r Iκ I +r D κ D +r PFR κ PFR
[0160] The peak-shaving charging and discharging powers of lithium battery energy storage at each decision point T in a day are The grid-connected energy cost of lithium battery energy storage is:
[0161]
[0162] In the formula is the on-grid electricity price at time T, is the electricity purchase price at time T.
[0163] Let the number of cells in the lithium battery energy storage be N c , the aging amount is The aging price of a cell per unit capacity is r Deg , then the aging cost of lithium battery energy storage when providing grid-building services is:
[0164]
[0165] In the formula The piecewise fitting aging characteristic function in step 1) is given according to the state of charge and power at the Tth moment.
[0166] The total economic benefits of lithium battery energy storage when providing grid-building services s for:
[0167] z s =z GFS +z G -z Deg
[0168] The frequency response power of lithium battery storage participating in the grid-building service at time t is The real-time grid-connected power of lithium battery energy storage is Given by the swing equation, It is composed of lithium battery energy storage peak-shaving power and frequency response power:
[0169]
[0170] Let the state of charge at the current decision point be The feasible domains of charging and discharging power output are determined as follows: There is an efficiency difference between the internal output power of lithium battery energy storage and the real-time grid-connected power. Let the efficiency of the lithium battery energy storage discharge side be η b , the charging efficiency is η c In the Tth decision step, the power of lithium battery energy storage should meet the following conditions:
[0171]
[0172] Where α ω represents the confidence level, and Pr{·} represents the probability. Furthermore, since The linear characteristics of the above formula can be converted into the power constraint of lithium battery energy storage:
[0173]
[0174] In the formula Indicates that within the Tth decision step α that constitutes the probability distribution ω Quantile.
[0175] The part is the power of lithium battery energy storage grid-connected discharge, let The part is the power of lithium battery energy storage grid charging, let The internal output charging power of the lithium battery energy storage is Discharge power They are:
[0176]
[0177] Based on the law of conservation of energy, the power values on different time scales are converted. At the Tth moment, the charging and discharging power of the lithium battery energy storage is the average power of the second-level sampling points within the current decision step, that is: Let the charging power sequence vector be P b,c ,Right now Similarly, let the discharge power sequence vector be P b,d , the state of charge sequence vector is C S The lithium battery energy storage charge state sequence can be calculated from the power sequence vector, as shown in the following formula:
[0178]
[0179] In the formula is the initial charge state, A0, A CS , A P is the state of charge update matrix, which can be given by the power-current characteristic coefficient of the electrochemical model. Then the state of charge constraint when lithium battery energy storage provides grid-building services is:
[0180]
[0181] In some embodiments, step 4) specifically includes:
[0182] Let the uncertainty set of power grid frequency data be Ω={ω1,ω2,…,ω n},in represents a frequency data sequence vector for one day. Let the network-based scheduling instruction set κ = {κ PFR ,κ D ,κ I}, the peak load power sequence vector is P G ,Right now in The economic benefits of lithium battery storage under grid uncertainty can be expressed as a function of κ:
[0183] z s (κ)=z GFs (κ)+E ω∈Ω (z G (κ,P G ,ω)-z Deg (κ,P G ,ω))
[0184] Where E ω∈Ω (·) means to find the expectation of the (·) part, z GFS , z G , z Deg is the lithium battery energy storage benefit function given in step 3). According to the above lithium battery energy storage benefit function and Benders algorithm, the grid-based lithium battery energy storage scheduling model can be divided into a main problem (MP) and multiple sub-problems (SP), where the initial MP is as follows:
[0185]
[0186] 0≤κ≤κ max
[0187] In the formula In all uncertain scenarios α that constitutes the probability distribution ω Quantile. is the static lithium battery energy storage power boundary, κ max is the maximum value of the network scheduling instruction. Solving the initial MP problem, we can get the optimal solution, denoted as κ * .
[0188] When the optimal solution of the main problem is obtained, for a certain ω i The SP under ∈Ω is expressed as:
[0189]
[0190] Step 3) Power Constraint
[0191] Step 3) State of Charge Constraints
[0192] For the SP problem, firstly, *Also regarded as the decision variable κ, the constraints can be transformed into a standard form through optimization problem modeling tools, as shown in the following formula:
[0193]
[0194] Where I is the unit variable, H κ , H G is the partial multiplier of the constraint matrix, and h is the rearranged right-hand side term. Solve for all ω i ∈Ω, The SP problem under i After solving the SP problem under ∈Ω, if the SP problem is infeasible, obtain the polar ray Add the following feasible cuts to the Benders cut set Γ:
[0195]
[0196] Among them, h i is the right-hand side term of the constraint condition in the i-th scenario, is the partial multiplier of the constraint matrix in the i-th scenario.
[0197] If all SP problems have optimal solutions, let the average value of the optimal value of the objective function of each sub-problem be y * , then UB is updated to UB = -z GFS (κ * )+y * For all ω i Obtaining the extreme points of SP problem under ∈Ω Add the following optimal cuts to the Benders cut set Γ:
[0198]
[0199] Where γ i is the probability of the i-th scenario in Ω, and θ is an auxiliary variable. After solving the SP problem, the following MP problem with Benders cut sets is solved again:
[0200]
[0201] 0≤κ≤κ max
[0202] Benders cut set Γ
[0203] Let LB be updated to the optimal value of the MP problem containing the Benders cut set LB = -z GFS (κ * )+θ *. Set the convergence criterion to ∈. When |UB-LB|≤∈, the scheduling problem is considered to have an optimal solution. The solution is the optimal solution κ to the MP problem containing Benders cut sets at this time. * Otherwise, repeat the above steps and set the optimal solution κ * Bring it into each subproblem, solve the subproblem and update the Benders cut set, and solve the MP problem containing the Benders cut set again until the convergence criterion is met.
[0204] Determine the solution to the scheduling problem * After that, it is used as the grid configuration coefficient dispatch instruction κ. For a specific grid frequency scenario ω s , the optimal solution can be obtained by solving the sub-problem SP at this time And obtain the charging and discharging power of lithium battery energy storage at each time, forming the charging and discharging power operation curve P b,d , P b,c .
[0205] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.
[0206] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.
[0207] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.
[0208] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute the instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.
[0209] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
[0210] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.
[0211] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.
[0212] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.
Claims
1. A grid-type energy storage scheduling method considering the mechanism characteristics of lithium batteries, characterized in that: include: Obtain the mapping relationship between the aging and decay of lithium battery energy storage and the state quantity of the dispatching side; Determine the economic value coefficient of lithium battery energy storage to provide grid-building services; Determine the aging cost function when the lithium battery energy storage provides a networking service based on the mapping relationship, determine the networking service revenue function when the lithium battery energy storage provides a networking service based on the economic value coefficient, and determine the energy cost function and constraint conditions when the lithium battery energy storage provides a networking service; Based on the aging cost function, the grid-forming service revenue function and the energy cost function, the economic benefit function of the lithium battery energy storage under the uncertainty of the grid frequency is determined, and the constraints are considered to maximize the economic benefits of the lithium battery energy storage, and a grid-forming coefficient scheduling instruction set of the grid-forming lithium battery energy storage is obtained.
2. The method according to claim 1, characterized in that The method of obtaining the mapping relationship between the aging and decay amount of the lithium battery energy storage and the state amount of the dispatching side includes: Step S1: obtaining an electrochemical model of the lithium battery energy storage; Step S2: obtaining the state of charge level and constant current excitation through uniformly distributed sampling within the feasible domain; Step S3: for the electrochemical model, given the state of charge level, applying the constant current excitation within a preset decision step to achieve simulation of the electrochemical model; Step S4: when simulating the electrochemical model, the side reaction intensity at each sampling moment is obtained at a sampling step of seconds, and the aging decay amount related to the side reaction is calculated based on the side reaction intensity obtained by sampling according to the parameters related to the electrode material in the electrochemical model, wherein the side reaction intensity includes the solid electrolyte interface film growth rate, the lithium precipitation rate and the active material loss rate; Step S5: accumulating the aging decay amount related to the side reaction over the complete simulation length to obtain the aging decay amount on the decision step scale; Step S6: when simulating the electrochemical model, obtaining the output voltage at each sampling moment at a sampling step of seconds, and determining the charge and discharge power at a decision step scale based on the constant current excitation and the output voltage; Step S7: repeating steps S2-S6 to obtain multiple sets of simulation data, wherein each set of simulation data includes a state of charge level, a charge and discharge power, and an aging degradation amount; Step S8: With the state of charge level and the charge and discharge power as independent variables and the aging degradation amount as the dependent variable, based on the multiple groups of simulation data, a piecewise linearized convex hull is used for fitting to obtain a fitting formula.
3. The method according to claim 2, characterized in that The aging decline associated with the side effects is: t SEI ∝j SEI ,t LP ∝j LP ,t AM ∝j AM Among them, τ SEI is the aging decay caused by the growth of the solid electrolyte interface film, j SEI is the solid electrolyte interface film growth rate, τ LP is the aging decay caused by lithium deposition, j LP is the lithium deposition rate, τ AM is the aging decay due to loss of active material, j AM is the active material loss rate; The aging decay amount on the decision step size scale is: Among them, N t is the preset decision step size; If the constant current excitation amplitude I0 ≥ 0, the charging power P on the decision step scale B,c is 0, the discharge power P B,d P L , if I0<0, the charging power P on the decision step scale B,c P L , discharge power P B,d is 0, where Among them, V t is the output voltage at sampling time t; The fitting formula is expressed as: in, is the mth segment, expressed as: Among them, C S is the state of charge, is the coefficient obtained by plane fitting of the mth segment.
4. The method according to claim 1, characterized in that The economic value coefficient of determining the grid-building service provided by lithium battery energy storage includes: Obtain historical external power grid frequency data for lithium battery energy storage to provide grid-building services; Based on the historical external power grid frequency data at each moment in the preset time period at the second-level sampling step, determine the external power grid frequency deviation of the primary frequency regulation service, damping and inertia response at each moment in the preset time period; According to the frequency deviation of the external power grid within the preset time period, the economic value brought to the power grid when the lithium battery energy storage provides 1 unit of primary frequency regulation service, 1 unit of damping, and 1 unit of inertia is determined respectively, and the determined 1 unit of economic value is used as the economic value coefficient of the corresponding grid-building service.
5. The method according to claim 4, characterized in that The external grid frequency deviations of primary frequency regulation service, damping and inertia response at each moment in the preset time period are: in, is the external grid frequency deviation of the primary frequency regulation service response at time t, is the external grid frequency deviation of the damping response at time t, is the external grid frequency deviation of the inertia response at time t, ω t is the external power grid frequency data at each moment, ω0 is the rated value of the power grid frequency, ω s is the external grid frequency deviation of the primary frequency modulation response after equivalent low-pass filtering, expressed as: Among them, Δω t-i is the frequency difference ω outside the primary frequency modulation service response dead zone at time ti t -ω0 t s is the response time constant of the primary frequency modulation service, t c is the cutoff constant; The economic value of one unit of primary frequency regulation service is: Among them, | αr is 1-α of the frequency deviation distribution r Confidence interval, is the average on-grid electricity price, N T is the total number of decision points; The economic value of 1 unit damping is: The economic value of 1 unit of inertia is: Among them, r I,init The economic value associated with inertia starting is given by the slope obtained by fitting a linear function using the starting fuel cost calculated with the hot cavity starting of different traditional thermal power units as the independent variable and the inertia coefficient of the thermal power unit as the dependent variable.
6. The method according to any one of claims 3 or 5, characterized in that: The aging cost function is: Among them, r Deg is the aging price of battery cell per unit capacity, N c is the number of cells in the lithium battery energy storage, The fitting formula is given according to the state of charge and power at the Tth moment; The networking service revenue function is expressed as: z GFS =r I k I +r D k D +r PFR k PFR Among them, κ I , κ D , κ PFR The coefficient that provides inertia, damping, and primary frequency regulation services for lithium battery energy storage, κ I , κ D , κ PFR Construct a network-based scheduling instruction set κ, κ = {κ PFR ,κ D ,κ I }; The energy cost function is expressed as: in, is the on-grid electricity price at time T, is the electricity purchase price at time T, It is the peak-shaving charging and discharging power of lithium battery energy storage at each decision point T in a day.
7. The method according to claim 1, characterized in that The constraints include state of charge constraints, power opportunity constraints and power constraints. The power opportunity constraints are expressed as: Among them, Pr{·} represents the probability, N t is the preset decision step size; according to The linear characteristics of the power opportunity constraint are converted into the power constraint, and the power constraint is expressed as: in, is the grid-connected power of lithium battery energy storage within the Tth decision step, is the real-time grid-connected power of lithium battery energy storage within the Tth decision step α that constitutes the probability distribution ω Quantile, α ω is the confidence level, η d is the discharge efficiency of lithium battery energy storage, η c The efficiency of the charging side of lithium battery energy storage, is the state of charge at the current decision point Determine the feasible domain of charging and discharging power output, and the real-time grid-connected power of lithium battery energy storage It is expressed as: in, The frequency response power of lithium battery storage participating in the grid-building service at time t, is the discharge power of lithium battery energy storage participating in peak regulation at time t, is the charging power of lithium battery energy storage participating in peak load regulation at time t; The state of charge constraint is expressed as: in, It is the lower limit of the lithium battery energy storage charge state. is the lithium battery energy storage charge state at the Tth decision step, is the upper limit of the lithium battery energy storage charge state, and the lithium battery energy storage charge state sequence vector C S for: Among them, P b,d is the discharge power sequence vector, P b,c is the charging power sequence vector, P b,c for P b,d for Among them, the power of lithium battery energy storage grid-connected discharge for Part of the power of lithium battery energy storage grid charging for part, It is the internal output discharge power of lithium battery energy storage. Output charging power for the internal lithium battery energy storage.
8. The method according to any one of claims 6 or 7, characterized in that: The economic benefit function of lithium battery energy storage under the uncertainty of grid frequency is: Where Ω is the uncertainty set of the power grid frequency data, Ω={ω1,ω2,…,ω n }, Represents a data sequence vector of the power grid frequency for one day, P G is the peak load power sequence vector, 9. The method according to claim 8, characterized in that The process of using the Benders algorithm to solve the grid-type coefficient scheduling instructions includes: Step S1: split the grid-connected lithium battery energy storage scheduling model into a main problem MP considering economic benefits and multiple sub-problems SP considering aging costs, energy costs and constraints corresponding to multiple scenarios under uncertain grid frequencies; Step S2: Solve the initial MP to obtain the optimal solution of MP; Step S3: Based on the optimal solution of MP, determine all corresponding ω i ∈Ω, and bring the optimal solution of MP into the determined value, and transform the constraints into standard form constraints; Step S4: Solve for all ω i When solving the SP problem under ∈Ω, if the current SP problem is not feasible, obtain the corresponding polar ray Adding polar rays to Benders cut sets The corresponding feasible cut, if all SP problems have optimal solutions, is based on the average value y of the optimal solutions of each SP problem * Update the upper bound UB and obtain all ω i The extreme points of SP problem under ∈Ω Adding vertices to the Benders cut set The corresponding optimal cut; Step S5: Solve the MP containing the Benders cut set, obtain the optimal solution of the MP, and update the lower bound LB to the optimal value of the MP containing the Benders cut set; Step S6: Repeat steps S3-S5 until |UB-LB|≤∈, and take the optimal solution of the MP problem containing the Benders cut set as the optimal solution of the scheduling problem, and use it as the scheduling instruction of the network configuration coefficient. s , bring in the grid-forming coefficient scheduling instruction as the determined value, solve the SP corresponding to the scenario to obtain the optimal solution, and obtain the charging and discharging power of the lithium battery energy storage at each moment to form a charging and discharging power operation curve.
10. The method according to claim 9, characterized in that The main problem MP is expressed as: in, For all uncertain scenarios α that constitutes the probability distribution ω Quantile; κ max It is the maximum value of the network scheduling instruction; For ω i The subproblem SP under ∈Ω is expressed as: st power constraint State of Charge Constraints Among them, κ * is the optimal solution of MP; The constraints in standard form are expressed as: Among them, I is the unit variable, H k , H G is the partial multiplier of the constraint matrix, and h is the rearranged right-hand side term; If the SP problem currently being solved is infeasible, the polar ray added to the Benders cut set Γ The corresponding feasible cut is: Among them, h i is the right-hand side term of the constraint condition in the i-th scenario, is the partial multiplier of the constraint matrix in the i-th scenario; The upper bound UB is updated to: UB6-z GFS (κ * )+y * If all SP problems have optimal solutions, the poles added to the Benders cut set Γ The corresponding optimal cut is: Among them, θ is an auxiliary variable, γ i is the probability of the i-th scenario in Ω; The MP containing the Benders cut set is: 0≤κ≤κ max Benders cut set Γ The lower bound LB is updated to: LB=-z GFS (κ * )+θ * Among them, θ * The optimal auxiliary variables obtained for solving MP with Benders cut sets.
Citation Information
Cited By
System safety assessment method and system for energy storage power station containing network-forming type energy storage converter
CN120975381A
