Network configuration and dispatching method for energy storage system based on frequency safety capacity margin

Abstract

This invention discloses a configuration and scheduling method for grid-based energy storage systems based on frequency safety capacity margin, belonging to the field of microgrid optimization technology. The method includes: obtaining the network topology of an islanded microgrid; establishing a frequency safety capacity margin model, and obtaining the dispatchable power and dispatchable energy of each grid-based energy storage system in the islanded microgrid based on the network topology and the frequency safety capacity margin model; establishing an optimized scheduling model based on the dispatchable power and dispatchable energy of each grid-based energy storage system in the islanded microgrid; simplifying the AC power flow constraints in the optimized scheduling model using convex relaxation processing; and solving the optimized scheduling model after convex relaxation processing based on binary selection variables and a commercial solver to obtain the configuration and scheduling results of the grid-based energy storage system. This method ensures sufficient inertia support and primary frequency regulation support capability after disturbances, and exhibits high solution efficiency and stability.

Description

Technical Field

[0001] This invention belongs to the field of microgrid optimization technology, specifically relating to a method for configuring and scheduling grid-type energy storage systems based on frequency safety capacity margin. Background Technology

[0002] In isolated microgrids where wind power, photovoltaics, and other new energy sources account for a high proportion of electricity supply, the microgrid operates independently from the main grid and cannot rely on external grids for frequency support. When the power of new energy units (such as wind turbines and solar turbines) fluctuates, load changes suddenly, or synchronous generators shut down, the isolated microgrid is prone to active power imbalance, which can lead to problems such as excessive frequency change rate and frequency deviation exceeding limits. This can affect normal power consumption on the user side and, in severe cases, cause local power outages or the shutdown of the isolated microgrid.

[0003] Grid-based energy storage systems can actively establish voltage and frequency references and provide inertia and primary frequency regulation support after disturbances occur. Therefore, during the capacity configuration phase of grid-based energy storage systems, their contribution to the system's inertia and frequency support capabilities needs to be considered. However, in actual operation, grid-based energy storage systems not only undertake frequency support tasks but also need to participate in economic dispatch, renewable energy consumption, and power balance.

[0004] Existing methods typically only set inertia constraints and frequency change rate constraints during the capacity configuration phase of grid-based energy storage systems to ensure that their rated capacity has a certain frequency support capability. However, they do not reflect the operational requirements for undertaking inertia support and primary frequency regulation tasks during the operation and scheduling phase. Since inertia support and primary frequency regulation mainly occur on a second-level time scale, while energy storage scheduling usually takes place on an hour-level time scale, the relevant frequency support requirements are difficult to directly incorporate into scheduling constraints, making it difficult to quantify the dispatchable power and dispatchable energy of grid-based energy storage systems. Therefore, even if grid-based energy storage meets frequency security requirements on the configuration side, actual scheduling may still excessively utilize its power and energy, making it difficult for it to provide the expected inertia support and primary frequency regulation support when disturbances occur.

[0005] Given that existing technologies cannot guarantee that grid-type energy storage systems have sufficient inertia support and primary frequency regulation capability after disturbances, a configuration and scheduling method for grid-type energy storage systems based on frequency safety capacity margin is proposed. Summary of the Invention

[0006] The purpose of this invention is to address the above-mentioned problems by proposing a configuration and scheduling method for grid-type energy storage systems based on frequency safety capacity margin. This method can ensure sufficient inertia support and primary frequency regulation support capability after disturbances, and has high solution efficiency and stability.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0008] The proposed method for configuring and scheduling a grid-type energy storage system based on frequency security capacity margin includes the following steps:

[0009] S1. Obtain the network topology of the isolated microgrid;

[0010] S2. Establish a frequency security capacity margin model, and based on the network topology and the frequency security capacity margin model, obtain the dispatchable power and dispatchable energy of each grid-type energy storage system in the islanded microgrid. Then, for each time period... Next node The dispatchable power of a grid-type energy storage system is the node The rated power of the grid-type energy storage system minus the time period Next node The sum of the inertia-supported power and primary frequency regulation-supported power of the grid-type energy storage system, and the dispatchable energy of the grid-type energy storage system at node i during time period t are the values ​​of node i. The rated energy capacity of the grid-type energy storage system minus the time period Next node The reserved energy capacity margin of the grid-type energy storage system , , This is a set of hourly scheduling time periods. Represents the set of nodes in a grid-type energy storage system, time period Next node The inertia support power, primary frequency regulation support power, and reserved energy capacity margin of the grid-type energy storage system are all obtained from the frequency safety capacity margin model.

[0011] S3. Establish an optimal dispatch model based on the dispatchable power and dispatchable energy of each grid-type energy storage system in the isolated microgrid, where:

[0012] 1) Objective function: Minimize the sum of investment cost, operation and maintenance cost, and operation and dispatch cost of all grid-type energy storage systems in the isolated microgrid;

[0013] 2) Constraints: These include configuration constraints, inertia support constraints, frequency security constraints, scheduling constraints, power balance constraints, unit power constraints, voltage constraints, and AC power flow constraints.

[0014] S4. Use convex relaxation to simplify and optimize the AC power flow constraints in the scheduling model.

[0015] S5. Solve the optimized scheduling model after convex relaxation based on binary selection variables and commercial solvers to obtain the configuration and scheduling results of the grid-type energy storage system.

[0016] Preferably, the frequency security capacity margin model is constructed as follows:

[0017] S21. Obtain the disturbance frequency response model of the isolated microgrid;

[0018] S22. Calculate the time period based on the disturbance frequency response model. Inertia support power of all grid-based energy storage systems in an isolated microgrid ;

[0019] S23. Calculate the time period based on the disturbance frequency response model. Primary frequency regulation support power of all grid-type energy storage systems in the isolated microgrid ;

[0020] S24, According to time period A frequency safety capacity margin model is established for the inertia-supported power and primary frequency regulation-supported power of all grid-type energy storage systems in the isolated microgrid. The formula for the frequency safety capacity margin model is as follows:

[0021]

[0022]

[0023] in, Indicates time period Reserved power capacity margin for all grid-type energy storage systems in isolated microgrids; Indicates time period Reserved energy capacity margin for all grid-type energy storage systems in isolated microgrids; Indicates the duration of inertial support, in hours; Indicates the start time of a frequency modulation, in hours; This indicates the time when a single frequency modulation reaches its rated support power, expressed in hours. Indicates the end time of a frequency modulation session, in hours; This indicates the discharge efficiency of a grid-type energy storage system.

[0024] Preferably, the dispatchable power and dispatchable energy of each grid-type energy storage system in an islanded microgrid also satisfy the following conditions:

[0025] 1) Time period The sum of the inertia support power of all nodes in the grid-type energy storage system is equal to the time period. Inertia support power of all grid-based energy storage systems in an isolated microgrid, time period The sum of the primary frequency regulation support power of all nodes in the grid-type energy storage system is equal to the time period. Primary frequency regulation support power of all grid-type energy storage systems in an isolated microgrid;

[0026] 2) Time period The sum of the reserved energy capacity margins of all nodes in the grid-type energy storage system equals the time period. Reserved energy capacity margin for all grid-type energy storage systems in the isolated microgrid.

[0027] Preferably, the time period is calculated based on the disturbance frequency response model. Primary frequency regulation support power of all grid-type energy storage systems in the isolated microgrid The details are as follows:

[0028] S231, Calculation Period The frequency deviation below is the time period. Active power disturbance of isolated microgrid minus time period Divide the primary frequency regulation support power of all grid-connected energy storage systems in the isolated microgrid by the time period. The load damping coefficient under the condition;

[0029] S232. Establish frequency deviation constraints, i.e., time period The frequency deviation is less than or equal to the maximum allowable frequency deviation of the islanded microgrid.

[0030] S233. Substitute the frequency deviation constraint into the formula for the frequency deviation under the corresponding time period to obtain the time period. Primary frequency regulation support power of all grid-type energy storage systems in an isolated microgrid.

[0031] Preferably, the scheduling constraints are as follows:

[0032] 1) Time period Next node The energy capacity of a grid-type energy storage system is equal to the time period. Next node The energy capacity of the grid-type energy storage system plus its charging efficiency and time period Next node The product of the charging power of the grid-type energy storage system and the time interval between adjacent time periods, minus the reciprocal of the discharge efficiency of the grid-type energy storage system, and the time period. Next node The product of the discharge power of the grid-type energy storage system and the time interval between adjacent time periods; and also greater than or equal to the time interval. Next node The reserved energy capacity margin of the grid-type energy storage system is less than or equal to that of the node. The rated energy capacity of a grid-type energy storage system;

[0033] 2) Time period Next node The charging power of the grid-type energy storage system is greater than or equal to 0 and less than or equal to the time period. Next node The charge and discharge status of grid-type energy storage systems The product of the rated power of the grid-type energy storage system, And it is greater than or equal to 0 and less than or equal to the time period. Next node The minimum of the maximum charging power and dispatchable power of the grid-type energy storage system;

[0034] 3) Time period Next node The discharge power of the grid-type energy storage system is greater than or equal to 0 and less than or equal to The product of the rated power of the grid-type energy storage system and the rated power of the grid-type energy storage system; and also greater than or equal to 0 and less than or equal to the time period. Next node The minimum of the following three values ​​for a grid-type energy storage system: the ratio of dispatchable energy to the time interval between adjacent periods, the maximum discharge power, and the dispatchable power.

[0035] Preferably, convex relaxation is used to simplify the AC power flow constraints in the optimization scheduling model, as follows:

[0036] S41, Time Period Acquisition The complex voltage vector of the nodes below;

[0037] S42, according to time period The construction period of the node complex voltage vector Voltage outer product matrix under;

[0038] S43, Time-based Solving the voltage outer product matrix under the following time period Next node The active and reactive power injections are calculated, and a second-order cone constraint is introduced.

[0039] Preferably, the second-order cone constraint simultaneously satisfies the following conditions:

[0040] 1) Time period Next node With nodes The L2 norm of the voltage cross product multiplied by 2 is less than or equal to the time interval. Next node With nodes The voltage cross product and time period Next node With nodes The sum of the voltage cross products between them;

[0041] 2) Time period Next node With nodes The L2 norm of the imaginary part of the voltage cross product between the two time intervals is less than or equal to the time interval. Next node With nodes The voltage cross product and time period Next node With nodes The sum of the voltage cross products between them;

[0042] 3) Time period Next node With nodes The L2 norm of the voltage cross product between time periods Next node With nodes The L2 norm of the difference in the voltage cross products between the time intervals is less than or equal to the time interval. Next node With nodes The voltage cross product and time period Next node With nodes The sum of the voltage cross products between them.

[0043] Preferably, the optimal scheduling model after convex relaxation is solved based on binary selection variables and a commercial solver to obtain the configuration and scheduling results of the grid-type energy storage system, as follows:

[0044] S51. Exhaustively search for candidate battery module configuration combinations to form a set of candidate battery module configuration combinations, wherein the first... Nodes in the candidate battery module configuration combination The number of battery modules in a grid-type energy storage system , , Nodes in order The lower and upper limits of the number of battery modules in a grid-type energy storage system. , A set of nodes in a network topology. , Indicates the number of candidate battery module configuration combinations;

[0045] S52. Traverse the candidate battery module configuration combination set, that is, each time select a candidate battery module configuration combination from the candidate battery module configuration combination set based on the binary selection variable.

[0046] S53. Based on the selected candidate battery module configuration combination, use a commercial solver to solve the optimized scheduling model after convex relaxation processing, and calculate the time period. The inertia support power, primary frequency regulation support power, reserved energy capacity margin, charging power and discharging power of the grid-type energy storage system at each node are used as scheduling parameters.

[0047] S54. The candidate battery module configuration combination with the objective function of the optimized scheduling model after convex relaxation is minimized is taken as the final battery module configuration combination, and together with the scheduling parameters, it is used as the configuration and scheduling result of the grid-type energy storage system for the corresponding time period.

[0048] Preferably, time period The inertia support power, primary frequency regulation support power, and reserved energy capacity margin of the grid-type energy storage system at each node are calculated as follows:

[0049] S531, Calculate the first Nodes in the candidate battery module configuration combination Support allocation coefficient of grid-type energy storage system That is, the first Nodes in the candidate battery module configuration combination The number of battery modules in the grid-type energy storage system and the first The ratio of the sum of the number of battery modules in all nodes of the grid-type energy storage system in a candidate battery module configuration combination;

[0050] S532, according to the Nodes in the candidate battery module configuration combination Support allocation coefficient of grid-type energy storage system And corresponding binary selection variables, calculate time period Next node Inertia support power of grid-type energy storage systems Time period Next node Primary frequency regulation support power of grid-type energy storage system and time period Next node Reserved energy capacity margin of grid-type energy storage systems The formula is as follows;

[0051]

[0052]

[0053]

[0054] in, For time period The inertia support power of all grid-type energy storage systems in an isolated microgrid. For time period The primary frequency regulation support power of all grid-type energy storage systems in an isolated microgrid. Indicates time period Reserved energy capacity margin for all grid-type energy storage systems in isolated microgrids Indicates the first Binary selection variables for a combination of candidate battery module configurations.

[0055] Preferably, the commercial solver is one of the following: Gurobi commercial solver, CPLEX commercial solver, or MOSEK commercial solver.

[0056] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0057] (1) This invention can quantify the frequency support capacity required per hour for a grid-type energy storage system. By establishing a frequency safety capacity margin model, the frequency change rate constraint and frequency deviation constraint after disturbance are transformed into inertia support power demand and primary frequency regulation support power demand, respectively, thereby clarifying the reserved power capacity margin and reserved energy capacity margin required by the grid-type energy storage system in each time period. This avoids the use of all rated power and rated energy for conventional charging and discharging scheduling, ensuring that sufficient inertia support and primary frequency regulation support capabilities are still available after a disturbance occurs;

[0058] (2) This invention can embed the second-level frequency response process into the hour-level optimized scheduling to achieve synergistic optimization. The inertia support requirements and primary frequency regulation requirements in the second-level frequency response process are transformed into power capacity margin and energy capacity margin in the hour-level optimized scheduling model, and the boundaries of the dispatchable power and dispatchable energy of the grid-type energy storage system are further modified. In this way, the second-level frequency dynamic constraints can be embedded into the hour-level energy storage configuration and scheduling process in the form of capacity margin, and an optimized scheduling model of the grid-type energy storage system considering frequency security is constructed, so that the energy storage configuration capacity, normal charging and discharging scheduling and frequency support capability can be synergistically optimized.

[0059] (3) This invention can improve the solution efficiency of grid-type energy storage system configuration and scheduling. It adopts a combination of convex relaxation treatment and candidate battery module configuration combination screening to accelerate the solution, transforming the battery module quantity configuration problem of grid-type energy storage system into a candidate battery module configuration combination selection problem, and uses second-order cone relaxation to treat AC power flow constraints, thereby reducing the solution difficulty of the original mixed integer nonlinear model and quickly obtaining the configuration and scheduling results of grid-type energy storage system;

[0060] In summary, this invention can improve the solution efficiency and convergence stability of the optimized scheduling model while taking into account the scheduling economic benefits, frequency security, and power flow operation of islanded microgrids. It is especially suitable for islanded microgrids with a high proportion of new energy access, insufficient synchronization inertia, and high frequency security requirements. Attached Figure Description

[0061] Figure 1This is a flowchart of the configuration and scheduling method for a grid-type energy storage system based on frequency security capacity margin according to the present invention.

[0062] Figure 2 This is a schematic diagram illustrating the inertia support and primary frequency modulation support processes of the present invention;

[0063] Figure 3 This is a schematic diagram of the network topology of the islanded microgrid of the present invention;

[0064] Figure 4 This is a schematic diagram showing wind power, solar power, and load power on different typical days. Figure 4 (a) in the figure is a schematic diagram of wind power output on different typical days. Figure 4 (b) in the diagram is a schematic diagram of the photoelectric power on different typical days. Figure 4 (c) in the diagram is a schematic diagram of the load power on different typical days;

[0065] Figure 5 This is a schematic diagram showing the frequency change rate and frequency deviation results for different typical days in existing islanded microgrids. Figure 5 (a) in the figure is a schematic diagram showing the frequency change rate results of existing islanded microgrids on different typical days. Figure 5 (b) in the figure is a schematic diagram of the frequency deviation results of existing islanded microgrids on different typical days;

[0066] Figure 6 This is a schematic diagram showing the frequency change rate and frequency deviation results of the islanded microgrid on different typical days according to the present invention, wherein, Figure 6 (a) in the figure is a schematic diagram showing the frequency change rate results of the islanded microgrid of the present invention on different typical days. Figure 6 (b) in the figure is a schematic diagram of the frequency deviation results of the islanded microgrid of the present invention on different typical days;

[0067] Figure 7 This is a schematic diagram showing the optimal gap before and after ablation in the accelerated solution module for different typical days of the present invention, wherein, Figure 7 In the diagram, (a) represents the optimal gap in the method of this invention. Figure 7 In the figure, (b) represents the optimal gap after removing the accelerated solution module in the method of the present invention. Detailed Implementation

[0068] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0069] It should be noted that when a component is referred to as being "connected" to another component, it can be directly connected to the other component or there may be an intervening component. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application.

[0070] like Figures 1-7 As shown, the configuration and scheduling method for grid-based energy storage systems based on frequency security capacity margin includes the following steps:

[0071] S1. Obtain the network topology of the isolated microgrid.

[0072] The network topology includes a set of nodes. and route set Node set The access nodes of the generators are included in the generator node set. The access nodes of new energy generating units are included in the new energy generating unit node set. The access nodes of the grid-type energy storage system are included in the grid-type energy storage system node set. Line collection Each line in the diagram represents a transmission line between different nodes.

[0073] Node set Each node in the diagram serves as the electrical connection point for connecting generators, renewable energy units, and grid-connected energy storage systems within an isolated microgrid. For example, the connection node for a generator is the generator node, the connection node for a renewable energy unit is the renewable energy unit node, and the connection node for a grid-connected energy storage system is the grid-connected energy storage system node. The set of generator nodes is... The set of new energy unit nodes is The node set of a grid-type energy storage system is , , and Both are sets of nodes A subset of [the category], new energy generating units generally include wind turbines (used to collect wind power data) and photovoltaic units (used to collect photovoltaic data), while grid-type energy storage systems consist of several battery modules. [Line set] A transmission line refers to a transmission line between different nodes (such as a power transmission line). The network topology is used to define the network connections, device access locations, and operational boundaries in the subsequent optimization and scheduling model.

[0074] S2. Establish a frequency security capacity margin model, and based on the network topology and the frequency security capacity margin model, obtain the dispatchable power and dispatchable energy of each grid-type energy storage system in the islanded microgrid. Then, for each time period... Next node The dispatchable power of a grid-type energy storage system is the node The rated power of the grid-type energy storage system minus the time period Next node The sum of the inertia-supported power and primary frequency regulation-supported power of the grid-type energy storage system, and the dispatchable energy of the grid-type energy storage system at node i during time period t are the values ​​of node i. The rated energy capacity of the grid-type energy storage system minus the time period Next node The reserved energy capacity margin of the grid-type energy storage system , , This is a set of hourly scheduling time periods. Represents the set of nodes in a grid-type energy storage system, time period Next node The inertia support power, primary frequency regulation support power, and reserved energy capacity margin of the grid-type energy storage system are all obtained from the frequency safety capacity margin model.

[0075] In one embodiment, the frequency security capacity margin model is constructed as follows:

[0076] S21. Obtain the disturbance frequency response model of the isolated microgrid;

[0077] S22. Calculate the time period based on the disturbance frequency response model. Inertia support power of all grid-based energy storage systems in an isolated microgrid ;

[0078] S23. Calculate the time period based on the disturbance frequency response model. Primary frequency regulation support power of all grid-type energy storage systems in the isolated microgrid ;

[0079] S24, According to time period A frequency safety capacity margin model is established for the inertia-supported power and primary frequency regulation-supported power of all grid-type energy storage systems in the isolated microgrid. The formula for the frequency safety capacity margin model is as follows:

[0080]

[0081]

[0082] in, Indicates time period Reserved power capacity margin for all grid-type energy storage systems in isolated microgrids; Indicates time period Reserved energy capacity margin for all grid-type energy storage systems in isolated microgrids; Indicates the duration of inertial support, in hours; Indicates the start time of a frequency modulation, in hours; This indicates the time when a single frequency modulation reaches its rated support power, expressed in hours. Indicates the end time of a frequency modulation session, in hours; This indicates the discharge efficiency of a grid-type energy storage system.

[0083] In one embodiment, the dispatchable power and dispatchable energy of each grid-type energy storage system in an islanded microgrid also satisfy the following conditions:

[0084] 1) Time period The sum of the inertia support power of all nodes in the grid-type energy storage system is equal to the time period. Inertia support power of all grid-based energy storage systems in an isolated microgrid, time period The sum of the primary frequency regulation support power of all nodes in the grid-type energy storage system is equal to the time period. Primary frequency regulation support power of all grid-type energy storage systems in an isolated microgrid;

[0085] 2) Time period The sum of the reserved energy capacity margins of all nodes in the grid-type energy storage system equals the time period. Reserved energy capacity margin for all grid-type energy storage systems in the isolated microgrid.

[0086] In one embodiment, the time period is calculated based on the disturbance frequency response model. Primary frequency regulation support power of all grid-type energy storage systems in the isolated microgrid The details are as follows:

[0087] S231, Calculation Period The frequency deviation below is the time period. Active power disturbance of isolated microgrid minus time period Divide the primary frequency regulation support power of all grid-connected energy storage systems in the isolated microgrid by the time period. The load damping coefficient under the condition;

[0088] S232. Establish frequency deviation constraints, i.e., time period The frequency deviation is less than or equal to the maximum allowable frequency deviation of the islanded microgrid.

[0089] S233. Substitute the frequency deviation constraint into the formula for the frequency deviation under the corresponding time period to obtain the time period. Primary frequency regulation support power of all grid-type energy storage systems in an isolated microgrid.

[0090] Specifically, the frequency security capacity margin model is constructed as follows:

[0091] S21. Obtain the disturbance frequency response model of the islanded microgrid, as shown in the following formula:

[0092]

[0093] in,

[0094]

[0095]

[0096] in, Indicates time period The equivalent inertia of an islanded microgrid; Indicates time period Next The rate of frequency change of an islanded microgrid per second; Indicates time period Active power disturbance in an isolated microgrid; Indicates time period Inertia support power of all grid-type energy storage systems in an isolated microgrid; Indicates time period Primary frequency regulation support power of all grid-type energy storage systems in an isolated microgrid; Indicates time period The load damping coefficient under the condition; Indicates time period Next Frequency deviation of islanded microgrids per second; Indicates time period Next Frequency deviation of islanded microgrids per second; This represents the time interval between adjacent moments, such as the first moment. Second and the The time interval in seconds; This indicates the maximum permissible rate of frequency variation for an islanded microgrid; Represents absolute value; , This is a set of hourly scheduling time periods.

[0097] In this context, it is assumed that the islanded microgrid experiences corresponding disturbances at any given time (e.g., during time intervals). Active power disturbance Its frequency changes dynamically on a second-scale timescale, meaning there is a second-scale frequency response process after a disturbance. To embed second-scale frequency security requirements into an hourly-scale optimized scheduling model, it is necessary to first analyze the frequency response process after the disturbance to construct a disturbance frequency response model. The disturbance frequency response model describes the relationship between active power disturbance, equivalent inertia, inertia-supported power, primary frequency regulation support, and load damping. The power margin and capacity margin in the subsequent frequency security capacity margin model are determined based on this disturbance frequency response model. Let the second-scale time index for the corresponding time period be denoted as . ,Right now This represents the time index variable indicating the frequency response process within the corresponding time period, in seconds. Frequency deviation is calculated based on the rated frequency of the islanded microgrid. For example, if 50Hz is used as the rated frequency, the frequency deviation equals the actual frequency minus the rated frequency. To avoid a rapid drop or rise in frequency after a disturbance, the absolute value of the rate of frequency change should be less than or equal to the maximum allowable rate of frequency change for the islanded microgrid. .

[0098] S22. Calculate the time period based on the disturbance frequency response model. Inertia support power of all grid-based energy storage systems in an isolated microgrid The formula is as follows:

[0099] .

[0100] Among them, in the time period It is assumed that the power of all devices in the isolated microgrid (including generators, renewable energy units, and grid-connected energy storage systems) remains constant during this time period, so it is only necessary to know the time period. By measuring the power of the device below, you can know the power of all devices in the next time period (the following hour).

[0101] Time period The initial frequency change rate of the disturbance is mainly affected by the active power disturbance, equivalent inertia, and inertia-supported power; primary frequency regulation has not yet occurred. To ensure that the frequency change rate does not exceed the limit, based on the disturbance frequency response model and constrained by the frequency change rate (… This yields the inertia support power of all grid-based energy storage systems in the islanded microgrid during the corresponding time period. The larger the equivalent inertia of the islanded microgrid, the greater the disturbance power it can withstand. When the equivalent inertia is insufficient, the grid-based energy storage system needs to supplement the corresponding inertia support power.

[0102] S23. Calculate the time period based on the disturbance frequency response model. Primary frequency regulation support power of all grid-type energy storage systems in the isolated microgrid The details are as follows:

[0103] S231, Calculation Period Frequency deviation The formula is as follows:

[0104]

[0105] S232. Establish frequency deviation constraints, as shown in the following formula:

[0106]

[0107] in, This indicates the maximum permissible frequency deviation for an islanded microgrid;

[0108] S233. Substitute the frequency deviation constraint into the formula for the frequency deviation under the corresponding time period to obtain the time period. Primary frequency regulation support power of all grid-type energy storage systems in the isolated microgrid :

[0109] .

[0110] Among them, time period After the disturbance enters the quasi-steady-state stage, the frequency deviation is mainly determined by the active power disturbance, load damping, and primary frequency regulation support. To ensure that the frequency deviation after the disturbance does not exceed the limit, the time period... The frequency deviation is less than or equal to the maximum allowable frequency deviation of the islanded microgrid, and the frequency deviation constraint is converted into the primary frequency regulation support power of all grid-connected energy storage systems. Thus, the inertia support power demand in the initial stage of the disturbance and the primary frequency regulation support power demand in the quasi-steady-state stage are both converted into time-limited... The power requirements below.

[0111] S24, According to time period A frequency safety capacity margin model is established for the inertia-supported power and primary frequency regulation-supported power of all grid-type energy storage systems in the isolated microgrid, as shown in the following formula:

[0112]

[0113]

[0114] in, Indicates time period Reserved power capacity margin for all grid-type energy storage systems in isolated microgrids; Indicates time period Reserved energy capacity margin for all grid-type energy storage systems in isolated microgrids; Indicates the duration of inertial support, in hours; Indicates the start time of a frequency modulation, in hours; This indicates the time when a single frequency modulation reaches its rated support power, expressed in hours. Indicates the end time of a frequency modulation session, in hours; This indicates the discharge efficiency of a grid-type energy storage system.

[0115] Since the optimal scheduling of grid-based energy storage systems is typically measured on an hourly timescale, while the frequency response process occurs on a second-scale timescale after a disturbance, directly embedding the complete second-scale frequency response process into the hourly-scale optimal scheduling model would cause difficulties in coupling variables at different timescales and significantly increase model complexity. To overcome these problems, this invention establishes a frequency safety capacity margin model to determine the dispatchable power and dispatchable energy of grid-based energy storage systems. Specifically, the frequency safety capacity margin obtained from the frequency safety capacity margin model includes both power capacity margin and energy capacity margin. Based on the inertia support power demand and primary frequency regulation support power demand obtained from the second-scale frequency response process analysis, the reserved power capacity margin and reserved energy capacity margin required by all grid-based energy storage systems in each time period are calculated. The actual dispatchable power for optimal scheduling is equal to the rated power of all grid-based energy storage systems minus the reserved power capacity margin, and the dispatchable energy is equal to the rated energy capacity of all grid-based energy storage systems minus the reserved energy capacity margin. This yields the boundaries of the dispatchable power and dispatchable energy available for normal optimal scheduling of the grid-based energy storage system per hour. The durations of inertia support and primary frequency modulation support are equivalent durations calculated by converting the second-level time scale to an hour-level time scale, and are preset parameters that can be adjusted according to actual needs. The processes of inertia support and primary frequency modulation support are as follows: Figure 2 As shown, Figure 2 middle, This indicates the start time of a frequency modulation. This indicates the moment when the frequency modulation reaches the rated support power. This indicates the moment when inertia support ends. This indicates the end time of a frequency modulation session, with the unit being seconds.

[0116] The frequency safety capacity margin model transforms second-level frequency support requirements into hour-level power and energy capacity margins. The reserved power capacity margin represents the power that a grid-based energy storage system must reserve to meet the inertia support and primary frequency regulation support power requirements after a disturbance during a corresponding time period. In other words, in hourly optimized scheduling, the grid-based energy storage system cannot use all its rated power for normal charging and discharging; it needs to reserve a power capacity margin for frequency support after a disturbance. Similarly, the reserved energy capacity margin represents the energy that a grid-based energy storage system needs to reserve during the duration of inertia support and primary frequency regulation support. In other words, in hourly optimized scheduling, the grid-based energy storage system cannot use all its rated energy for normal charging and discharging; it must reserve a certain energy capacity to ensure that the frequency support process can be completed according to the required support power and duration after a disturbance occurs. The frequency safety capacity margin model allows second-level frequency response requirements to be entered into the hourly optimized scheduling model in the form of power and energy capacity margins, thereby overcoming the problem of inconsistency between the time scale of the frequency response process and the time scale of optimized scheduling.

[0117] Specifically, based on the network topology and frequency security capacity margin model, the dispatchable power and dispatchable energy of each grid-type energy storage system in an islanded microgrid are obtained, where:

[0118] 1) Time period Next node Dispatchable power of grid-type energy storage systems The calculation is as follows:

[0119]

[0120] and

[0121]

[0122]

[0123] in, Represents a node The rated power of the grid-type energy storage system; For time period Inertia support power of all grid-type energy storage systems in an isolated microgrid; For time period Primary frequency regulation support power of all grid-type energy storage systems in an isolated microgrid; For time period Next node The inertia-supported power of a grid-type energy storage system; For time period Next node The primary frequency regulation support power of the grid-type energy storage system; , This represents the set of nodes in a grid-type energy storage system.

[0124] 2) Dispatchable energy of the grid-type energy storage system at node i during time period t The calculation is as follows:

[0125]

[0126] and

[0127]

[0128] in, Indicates time period Reserved energy capacity margin for all grid-type energy storage systems in isolated microgrids; Represents a node The rated energy capacity of a grid-type energy storage system; Indicates time period Next node The reserved energy capacity margin of the grid-type energy storage system.

[0129] In this embodiment, in order to calculate the inertia support power and primary frequency regulation power of each grid-type energy storage system, it is preferable to allocate them according to a ratio, as follows:

[0130] Power capacity margins are allocated according to the proportion of the rated power of each grid-type energy storage system to the sum of the rated power of all grid-type energy storage systems, including inertia-supported power and primary frequency regulation-supported power:

[0131]

[0132]

[0133] in, Represents a node The rated power of the grid-type energy storage system.

[0134] The energy capacity margin is allocated according to the proportion of the rated energy capacity of each grid-type energy storage system to the sum of the rated energy capacities of all grid-type energy storage systems:

[0135]

[0136] in, Represents a node The rated energy capacity of a grid-type energy storage system; Represents a node The rated energy capacity of the grid-type energy storage system.

[0137] Optimize scheduling based on the dispatchable power and dispatchable energy of each grid-type energy storage system, and allocate frequency safety capacity margin to each grid-type energy storage system.

[0138] S3. Establish an optimal dispatch model based on the dispatchable power and dispatchable energy of each grid-type energy storage system in the islanded microgrid, where:

[0139] 1) Objective function: Minimize the sum of investment cost, operation and maintenance cost, and operation and dispatch cost of all grid-type energy storage systems in the isolated microgrid;

[0140] 2) Constraints: These include configuration constraints, inertia support constraints, frequency security constraints, scheduling constraints, power balance constraints, unit power constraints, voltage constraints, and AC power flow constraints.

[0141] In one embodiment, the scheduling constraints are as follows:

[0142] 1) Time period Next node The energy capacity of a grid-type energy storage system is equal to the time period. Next node The energy capacity of the grid-type energy storage system plus its charging efficiency and time period Next node The product of the charging power of the grid-type energy storage system and the time interval between adjacent time periods, minus the reciprocal of the discharge efficiency of the grid-type energy storage system, and the time period. Next node The product of the discharge power of the grid-type energy storage system and the time interval between adjacent time periods; and also greater than or equal to the time interval. Next node The reserved energy capacity margin of the grid-type energy storage system is less than or equal to that of the node. The rated energy capacity of a grid-type energy storage system;

[0143] 2) Time period Next node The charging power of the grid-type energy storage system is greater than or equal to 0 and less than or equal to the time period. Next node The charge and discharge status of grid-type energy storage systems The product of the rated power of the grid-type energy storage system, And it is greater than or equal to 0 and less than or equal to the time period. Next node The minimum of the maximum charging power and dispatchable power of the grid-type energy storage system;

[0144] 3) Time period Next node The discharge power of the grid-type energy storage system is greater than or equal to 0 and less than or equal to 0. The product of the rated power of the grid-type energy storage system and the rated power of the grid-type energy storage system; and also greater than or equal to 0 and less than or equal to the time period. Next node The minimum of the following three values ​​for a grid-type energy storage system: the ratio of dispatchable energy to the time interval between adjacent periods, the maximum discharge power, and the dispatchable power.

[0145] Specifically, the scheduling model is optimized and established as follows:

[0146] S31. Constructing the objective function :

[0147]

[0148]

[0149] in, This indicates taking the minimum value; This represents the total cost of an islanded microgrid; This represents the total investment cost of the entire grid-type energy storage system; This represents the total operation and maintenance cost of the entire grid-type energy storage system; This represents the operation and scheduling cost of an isolated microgrid.

[0150] Investment cost of all grid-type energy storage systems The calculation is as follows:

[0151]

[0152] in, This represents the power investment cost coefficient of a grid-type energy storage system; This represents the energy investment cost coefficient of a grid-type energy storage system; Represents a node The rated power of the grid-type energy storage system; Represents a node The rated energy capacity of a grid-type energy storage system; Indicates the discount rate; This indicates the lifespan of a grid-type energy storage system.

[0153] Operation and maintenance costs of all grid-type energy storage systems The calculation is as follows:

[0154]

[0155] in, This represents the operation and maintenance cost coefficient of a grid-type energy storage system.

[0156] Operation and scheduling costs of isolated microgrids The calculation is as follows:

[0157]

[0158] in, Indicates time period Operating costs of the generator; Indicates time period Reduce the cost of penalties for abandoning renewable energy sources; Indicates time period Reduce network loss costs; Indicates time period Lower voltage deviation penalty cost; Indicates time period Lower frequency deviation penalty cost.

[0159] Among them, time period Operating costs of generators The calculation is as follows:

[0160]

[0161] in, Represents a node The first operating cost coefficient of the generator; Represents a node The second operating cost coefficient of the generator; Represents a node The third operating cost coefficient of the generator; Indicates time period Next node The active power of the generator; Indicates time period Next node The generator's start / stop status, and , Representative node The generator is in a stopped state. Representative node The generator is running. , This is the set of generator nodes.

[0162] Time period Lowering the cost of penalties for curtailing renewable energy The calculation is as follows:

[0163]

[0164] in, This represents the penalty coefficient for curtailment of renewable energy. Indicates time period Next node The predicted power of the new energy generating units; Indicates time period Next node The actual power output of the new energy generating units; , This is a set of nodes for new energy generating units.

[0165] Time period Lower network loss cost The calculation is as follows:

[0166]

[0167] in, This represents the network loss cost coefficient; Indicates the line The resistance; Indicates time period Downline The active transmission power; Indicates time period Downline reactive power transmission; Indicates time period Downline The voltage amplitude at the starting node; , This is a set of routes.

[0168] Time period Lower voltage deviation penalty cost The calculation is as follows:

[0169]

[0170] in, This represents the voltage deviation penalty coefficient; Indicates time period Next node The voltage amplitude; Indicates the reference voltage; , It is a set of nodes.

[0171] Time period Lower frequency deviation penalty cost The calculation is as follows:

[0172]

[0173] in, This represents the frequency deviation penalty coefficient; Indicates time period The frequency deviation below.

[0174] The total cost of an islanded microgrid includes the investment cost of all grid-connected energy storage systems, the operation and maintenance cost of all grid-connected energy storage systems, and the operation and dispatch cost of the islanded microgrid. The operation and dispatch cost of the islanded microgrid also includes generator operation costs, renewable energy curtailment penalty costs, network loss costs, voltage deviation penalty costs, and frequency deviation penalty costs. The optimization objective of this invention is to collaboratively determine the configuration capacity and hourly dispatch scheme of the grid-connected energy storage system, while satisfying the constraints of the islanded microgrid (including configuration constraints, inertia support constraints, frequency security constraints, dispatch constraints, power balance constraints, generator power constraints, voltage constraints, and AC power flow constraints of the grid-connected energy storage system), so as to minimize the total cost of the islanded microgrid.

[0175] S32. Establish constraints as follows:

[0176] 1) Configuration constraints:

[0177] The energy storage batteries in a grid-based energy storage system adopt a modular configuration, meaning that each battery module is the smallest configuration unit. Each battery module has a fixed rated power and rated energy capacity. The rated power and rated energy capacity of a single grid-based energy storage system are determined by the number of battery modules connected. The configuration constraints satisfy the following formula:

[0178]

[0179]

[0180]

[0181] in, Represents a node The rated power of the grid-type energy storage system; Represents a node The rated energy capacity of a grid-type energy storage system; For nodes The number of battery modules in a grid-type energy storage system; Indicates the rated power of a single battery module; This indicates the rated energy capacity of a single battery module; Represents a node The lower limit of the number of battery modules in a grid-type energy storage system; Represents a node The upper limit on the number of battery modules in a grid-type energy storage system.

[0182] 2) Inertia support constraint:

[0183] Since grid-based energy storage systems need to participate in the inertia support of islanded microgrids, the equivalent inertia of the configured islanded microgrid should include the inertia of the generators and the equivalent inertia of the grid-based energy storage system.

[0184]

[0185] in, Indicates time period The equivalent inertia of an islanded microgrid; Represents a node The inertia time constant of the generator; Represents a node The rated energy capacity of the generator; Indicates time period Next node The generator's start / stop status; This represents the equivalent inertia time constant of a grid-type energy storage system.

[0186] To meet the frequency change rate requirement after disturbance, the equivalent inertia of an islanded microgrid satisfies:

[0187]

[0188] At the same time, the rate of change of frequency satisfies:

[0189]

[0190] in, This indicates the rated frequency of the islanded microgrid.

[0191] These constraints indicate that the configuration capacity of a grid-based energy storage system not only determines the dispatch boundary but also directly affects the security of the equivalent inertia and frequency change rate of an isolated microgrid.

[0192] 3) Frequency security constraints:

[0193] Frequency security constraints include the frequency support power of a grid-connected energy storage system at a single node, which is reserved to ensure frequency security requirements. Specifically, the power capacity margin cannot exceed the rated power of the corresponding node's grid-connected energy storage system, and power allocation must be performed proportionally, as shown in the following formula:

[0194]

[0195]

[0196]

[0197] in, For time period Next node The reserved power capacity margin of the grid-type energy storage system.

[0198] This formula indicates that the number of battery modules in a grid-type energy storage system must be sufficient to ensure that the rated power of the grid-type energy storage system is greater than or equal to the sum of the inertia support power and the primary frequency regulation support power of the grid-type energy storage system.

[0199] In terms of energy capacity, grid-based energy storage systems need to meet the energy capacity margin requirements supported by frequency, and need to complete energy distribution proportionally.

[0200]

[0201]

[0202] This formula indicates that the number of battery modules in a grid-type energy storage system must also be sufficient to ensure that the rated energy capacity of the grid-type energy storage system is greater than or equal to the energy capacity margin required for frequency support.

[0203] 4) Scheduling constraints:

[0204] The energy capacity state of a grid-type energy storage system satisfies:

[0205]

[0206] in, Indicates time period Next node The energy capacity of a grid-type energy storage system; Indicates time period Next node The energy capacity of a grid-type energy storage system; Indicates the charging efficiency of a grid-type energy storage system; Indicates the discharge efficiency of a grid-type energy storage system; Indicates time period Next node The charging power of the grid-type energy storage system; Indicates time period Next node The discharge power of the grid-type energy storage system; This indicates the time interval between adjacent time periods.

[0207] The charge / discharge state of a grid-type energy storage system satisfies:

[0208]

[0209]

[0210] in, Indicates time period Next node The charging and discharging states of the grid-type energy storage system, and ,when Time, period Next node The grid-type energy storage system is in a charging state; when Time, period Next node The grid-type energy storage system is in a discharge state; This refers to the rated power of a grid-type energy storage system.

[0211] After considering frequency safety capacity margin, the lower limit of the energy capacity state of a grid-type energy storage system should at least retain frequency support capacity margin:

[0212]

[0213] The charging power of grid-based energy storage systems should also be limited by both remaining capacity and dispatchable power.

[0214]

[0215] in, Indicates time period Next node The maximum charging power of the grid-type energy storage system is determined by the remaining capacity space. Decide.

[0216] The discharge power of a grid-type energy storage system should also be limited by the release power, dispatchable power, and dispatchable energy.

[0217]

[0218] in, Indicates time period Next node The maximum discharge power of the grid-type energy storage system.

[0219] To ensure sufficient power capacity margin for frequency support, the time period Next node Energy capacity of grid-type energy storage systems It cannot be fully illuminated; at least a certain period of time must be reserved. Next node Reserved energy capacity margin of grid-type energy storage systems Therefore, the release capacity is only Thus, the time period is obtained. Next node The maximum discharge power of the grid-type energy storage system.

[0220] As can be seen from step S2, the time period Next node The dispatchable power of the grid-type energy storage system is Time period Next node The dispatchable energy of the grid-type energy storage system is Therefore, the charging and discharging power and capacity status of a grid-type energy storage system during normal dispatch should be limited by the aforementioned dispatchable boundaries (dispatchable power and dispatchable energy).

[0221] 5) Power balance constraints:

[0222]

[0223]

[0224] in, Indicates time period Next node The active power injected; Indicates time period Next node The active power of the generator (this value is 0 if node n has no generator). Indicates time period Next node The actual power of the new energy generator unit (this value is 0 if node n has no new energy generator unit). Indicates time period Next node The discharge power of the grid-type energy storage system; Indicates time period Next node The charging power of the grid-type energy storage system (this value is 0 if node n does not have a grid-type energy storage system). Indicates time period Next node Active load; Indicates time period Next node The reactive power injected; Indicates time period The reactive power of the generator (this value is 0 if node n has no generator). Indicates time period Next node The reactive load.

[0225] 6) Unit power constraints:

[0226] Generator power constraints include active power constraints, reactive power constraints, and ramping power constraints.

[0227]

[0228]

[0229]

[0230] in, Indicates time period Next node The active power of the generator; Indicates time period Next node The active power of the generator; Indicates time period Next node The reactive power injected; This indicates the lower limit of the generator's active power. This indicates the upper limit of the generator's active power; This indicates the lower limit of the generator's reactive power. This indicates the upper limit of the generator's reactive power; This indicates the upper limit of the generator's ramping power.

[0231] Meanwhile, the power constraints of the new energy units are met:

[0232]

[0233] in, Indicates time period Next node The actual power output of the new energy generating units; Indicates time period Next node The predicted power of the new energy units.

[0234] 7) Voltage constraint:

[0235]

[0236] in, Indicates time period Next node The voltage amplitude; This indicates the lower voltage limit of the node. This indicates the upper limit of the node's voltage.

[0237] To ensure that the scheduling results of isolated microgrids meet the network power flow operation rules, AC power flow constraints need to be further established on the basis of node power balance constraints.

[0238] 8) Constraints on the flow of communication:

[0239] Time period Next node Active power injection and time period Next node reactive power injection They respectively satisfy:

[0240]

[0241] in, Indicates time period Next node The voltage amplitude; Indicates time period Next node The voltage amplitude; Indicates time period Next node With nodes The voltage phase angle difference between them; Represents a node With nodes The electrical conductance between the lines, Represents a node With nodes Line susceptance between; , , It is a set of nodes.

[0242] S4. Use convex relaxation to simplify the AC power flow constraints in the optimization scheduling model.

[0243] In one embodiment, convex relaxation is used to simplify the AC power flow constraints in the optimization scheduling model, as follows:

[0244] S41, Time Period Acquisition The node complex voltage vector;

[0245] S42, According to time period The construction period of the node complex voltage vector Voltage outer product matrix under;

[0246] S43, Time-based Solving the voltage outer product matrix under the following time period Next node The active and reactive power injections are calculated, and a second-order cone constraint is introduced.

[0247] In one embodiment, the second-order cone constraint simultaneously satisfies the following conditions:

[0248] 1) Time period Next node With nodes The L2 norm of the voltage cross product multiplied by 2 is less than or equal to the time interval. Next node With nodes The voltage cross product and time period Next node With nodes The sum of the voltage cross products between them;

[0249] 2) Time period Next node With nodes The L2 norm of the imaginary part of the voltage cross product between the two time intervals is less than or equal to the time interval. Next node With nodes The voltage cross product and time period Next node With nodes The sum of the voltage cross products between them;

[0250] 3) Time period Next node With nodes The L2 norm of the voltage cross product between time periods Next node With nodes The L2 norm of the difference in the voltage cross products between the time intervals is less than or equal to the time interval. Next node With nodes The voltage cross product and time period Next node With nodes The sum of the voltage cross products between them.

[0251] Specifically, the convex relaxation process is as follows:

[0252] S41, Set Time Period The node complex voltage vector for:

[0253]

[0254] in, Indicates time period Next node The voltage amplitude; Indicates time period Next node The voltage phase angle; Indicates the number of nodes in an islanded microgrid; Indicates transpose; It represents the imaginary unit.

[0255] S42, Construction Period Voltage outer product matrix :

[0256]

[0257] in, This indicates the conjugate transpose. The voltage outer product matrix is ​​used to transform the voltage magnitude product and phase angle difference terms in the AC power flow constraint equations into matrix element expressions, thus facilitating subsequent convex relaxation processing.

[0258] Time period Voltage outer product matrix The Middle Line number Column elements Represented as:

[0259]

[0260] in, express The real part; express The imaginary part, , .

[0261] S43, Time-based Voltage outer product matrix Solving time period Next node Active power injection and time period Next node reactive power injection And a second-order cone constraint is introduced, where:

[0262] 1) Time period Next node Active power injection and time period Next node reactive power injection The calculation is as follows:

[0263]

[0264]

[0265] Achieve time period Next node The active and reactive power injections are written in linear form.

[0266] 2) Second-order cone constraint, expressed as follows:

[0267]

[0268] in, For time period Voltage outer product matrix The Middle Line number Column elements, Indicates time period Voltage outer product matrix The Middle Line number The elements of the column, and ; Indicates time period Voltage outer product matrix The Middle Line number Column elements; Represents the L2 norm;

[0269] To further improve solution efficiency, convex relaxation is applied to the AC power flow constraints. Second-order cone constraints are used to relax the non-convex rank constraints of the voltage outer product matrix, allowing the AC power flow constraints to be embedded in the optimal scheduling model of the grid-type energy storage system in a second-order cone form. (Time period) Next node Active power injection Time period Next node reactive power injection Together with the second-order cone constraint, it serves as a simplified result of the AC power flow constraint. Through the above convex relaxation process, the originally non-convex AC power flow constraint is transformed into a more easily solvable convex relaxation form, thereby improving the model's solution efficiency and numerical stability.

[0270] S5. Solve the optimized scheduling model after convex relaxation based on binary selection variables and commercial solvers to obtain the configuration and scheduling results of the grid-type energy storage system.

[0271] In one embodiment, the optimized scheduling model after convex relaxation is solved based on binary selection variables and a commercial solver to obtain the configuration and scheduling results of the grid-type energy storage system, as follows:

[0272] S51. Exhaustively search for candidate battery module configuration combinations to form a set of candidate battery module configuration combinations, wherein the first... Nodes in the candidate battery module configuration combination The number of battery modules in a grid-type energy storage system , , Nodes in order The lower and upper limits of the number of battery modules in a grid-type energy storage system. , A set of nodes in a network topology. , Indicates the number of candidate battery module configuration combinations;

[0273] S52. Traverse the candidate battery module configuration combination set, that is, each time select a candidate battery module configuration combination from the candidate battery module configuration combination set based on the binary selection variable.

[0274] S53. Based on the selected candidate battery module configuration combination, use a commercial solver to solve the optimized scheduling model after convex relaxation processing, and calculate the time period. The inertia support power, primary frequency regulation support power, reserved energy capacity margin, charging power and discharging power of the grid-type energy storage system at each node are used as scheduling parameters.

[0275] S54. The candidate battery module configuration combination with the objective function of the optimized scheduling model after convex relaxation is minimized is taken as the final battery module configuration combination, and together with the scheduling parameters, it is used as the configuration and scheduling result of the grid-type energy storage system for the corresponding time period.

[0276] In one embodiment, time period The inertia support power, primary frequency regulation support power, and reserved energy capacity margin of the grid-type energy storage system at each node are calculated as follows:

[0277] S531, Calculate the first Nodes in the candidate battery module configuration combination Support allocation coefficient of grid-type energy storage system That is, the first Nodes in the candidate battery module configuration combination The number of battery modules in the grid-type energy storage system and the first The ratio of the sum of the number of battery modules in all nodes of the grid-type energy storage system in a candidate battery module configuration combination;

[0278] S532, according to the Nodes in the candidate battery module configuration combination Support allocation coefficient of grid-type energy storage system And corresponding binary selection variables, calculate time period Next node Inertia support power of grid-type energy storage systems Time period Next node Primary frequency regulation support power of grid-type energy storage system and time period Next node Reserved energy capacity margin of grid-type energy storage systems The formula is as follows;

[0279]

[0280]

[0281]

[0282] in, For time period The inertia support power of all grid-type energy storage systems in an isolated microgrid. For time period The primary frequency regulation support power of all grid-type energy storage systems in an isolated microgrid. Indicates time period Reserved energy capacity margin for all grid-type energy storage systems in isolated microgrids Indicates the first Binary selection variables for a combination of candidate battery module configurations.

[0283] In one embodiment, the commercial solver is one of the Gurobi commercial solver, the CPLEX commercial solver, or the MOSEK commercial solver.

[0284] Specifically, the configuration and scheduling results of the grid-type energy storage system are obtained as follows:

[0285] S51. Exhaustively search for candidate battery module configuration combinations to form a set of candidate battery module configuration combinations. , means as follows:

[0286]

[0287] Among them, the Candidate battery module configuration combinations , means as follows:

[0288]

[0289]

[0290] in, Indicates the first Nodes in the candidate battery module configuration combination The number of battery modules in a grid-type energy storage system; Represents a node The lower limit of the number of battery modules in a grid-type energy storage system; Represents a node The upper limit on the number of battery modules in a grid-type energy storage system; , This indicates the number of candidate battery module configuration combinations. All candidate battery module configuration combinations are generated based on the upper and lower limits of the number of battery modules corresponding to candidate nodes in a grid-type energy storage system.

[0291] S52. Introduce binary selection variables to select candidate battery module configuration combinations from the candidate battery module configuration combination set, as follows:

[0292] Selecting the first based on binary selection variables The candidate battery module configuration combinations are selected as the current candidate battery module configuration combinations. :

[0293]

[0294] in,

[0295]

[0296] Among them, binary selection variables , Indicates the first The binary selection variables for the combination of candidate battery module configurations, i.e. , Indicates the selection of the first A combination of candidate battery module configurations; Indicates not to select the first option A combination of candidate battery module configurations.

[0297] Furthermore, nodes Rated power of grid-type energy storage system and nodes Rated energy capacity of grid-type energy storage system The calculation is as follows:

[0298]

[0299]

[0300] in,

[0301]

[0302] in, For nodes The final number of battery modules in a grid-type energy storage system.

[0303] To select the final solution from all candidate battery module configuration combinations, a binary selection variable is introduced, and a unique candidate battery module configuration combination is ultimately selected as the final candidate battery module configuration combination based on the constraints of the binary selection variable.

[0304] S53. Based on the selected candidate battery module configuration combination, use a commercial solver to solve the optimized scheduling model after convex relaxation processing, and calculate the time period. The inertia support power, primary frequency regulation support power, reserved energy capacity margin, charging power, and discharging power of the grid-type energy storage system at each node are used as scheduling parameters.

[0305] 5.3.1) Calculate the first... Nodes in the candidate battery module configuration combination Support allocation coefficient of grid-type energy storage system The formula is as follows:

[0306]

[0307] in, Indicates the first Nodes in the candidate battery module configuration combination The number of battery modules in a grid-type energy storage system.

[0308] Calculation period Next node Inertia support power of grid-type energy storage systems Time period Next node Primary frequency regulation support power of grid-type energy storage system Time period Next node Reserved energy capacity margin of grid-type energy storage systems The formula is as follows:

[0309]

[0310]

[0311]

[0312] in, For time period The inertia support power of all grid-type energy storage systems in an isolated microgrid. For time period The primary frequency regulation support power of all grid-type energy storage systems in an isolated microgrid. Indicates time period Reserved energy capacity margin for all grid-type energy storage systems in the isolated microgrid.

[0313] Because of the time period Primary frequency regulation support power of all grid-type energy storage systems in the isolated microgrid To optimize the variables, then It is the product of a continuous variable and a binary choice variable, and is a non-linear term.

[0314] Furthermore, to accelerate the solution, the third... The candidate battery module configuration combinations in the time period Auxiliary variables under :

[0315]

[0316] and

[0317]

[0318]

[0319] in, It is the maximum value of the primary frequency regulation support power, and is less than or equal to the rated power of the grid-type energy storage system.

[0320] Therefore, time period Next node Primary frequency regulation support power of grid-type energy storage system The calculation is as follows:

[0321]

[0322] Through the above processing, the original content The bilinear assignment relation is transformed into a mixed-integer linear constraint, which can further accelerate the solution.

[0323] 5.3.2) For each selected candidate battery module configuration combination, time period The charging and discharging power of the grid-type energy storage system at each node is obtained by a commercial solver through joint optimization, under the constraints of the optimized scheduling model after convex relaxation (especially the scheduling constraints and power balance constraints), with the objective of minimizing the objective function of the optimized scheduling model after convex relaxation.

[0324] S54. The candidate battery module configuration combination with the objective function of the optimized scheduling model after convex relaxation is minimized is taken as the final battery module configuration combination, and together with the scheduling parameters, it is used as the configuration and scheduling result of the grid-type energy storage system for the corresponding time period.

[0325] Specifically, the candidate battery module configuration combination that minimizes the objective function of the optimized scheduling model after convex relaxation processing corresponding to each candidate battery module configuration combination is taken as the final battery module configuration combination. Based on the final battery module configuration combination, the number of battery modules in the grid-type energy storage system at each node is determined, which further determines the corresponding rated power and rated energy capacity. Based on the inertia support power, primary frequency regulation support power, and reserved energy capacity margin of the grid-type energy storage system at each node in each time period, its dispatchable power and dispatchable energy can be determined as the power and energy capacity boundaries for hourly energy storage charging and discharging. Under the above power and energy capacity boundaries, the commercial solver further solves the optimized scheduling model to obtain the charging power, discharging power, and energy capacity of the grid-type energy storage system at each node in each time period, thereby completing the capacity configuration and hourly scheduling of the grid-type energy storage system.

[0326] Because the number of battery modules in a grid-based energy storage system is an integer variable, and the optimization scheduling model also includes configuration constraints, inertia support constraints, frequency security constraints, scheduling constraints, power balance constraints, generator power constraints, voltage constraints, and AC power flow constraints, the optimization scheduling model exhibits strong mixed-integer nonlinear characteristics, resulting in long solution time and poor convergence stability. To improve solution efficiency, this invention employs an accelerated solution process (referred to as the accelerated solution module) that combines convex relaxation processing with candidate battery module configuration combination screening. This obtains the capacity configuration and hourly scheduling scheme of the grid-based energy storage system that satisfies frequency security and minimizes the total cost of the islanded microgrid, serving as the final result of the collaborative configuration and scheduling of the grid-based energy storage system. Specifically, it first generates all feasible candidate battery module configuration combinations based on the upper and lower limits of the number of battery modules in the grid-based energy storage system at each node. Then, it selects the final configuration scheme from the candidate battery module configuration combinations using binary selection variables, thereby transforming the original node-by-node integer configuration problem into a candidate battery module configuration combination selection problem.

[0327] To verify the method of this invention, this embodiment conducts an analysis based on an IEEE 14-node islanded microgrid. A schematic diagram of the network topology of the islanded microgrid is shown below. Figure 3 As shown, this is used to demonstrate the access locations and line connections of various devices (generators, grid-connected energy storage systems, new energy units, and loads). Figure 3 In the diagram, 1-14 represent node numbers; thick black horizontal lines represent node buses; thin black lines represent transmission lines between nodes; and black arrows represent loads. The voltage amplitude range of each node is... The rated power of a single battery module in a grid-type energy storage system is... Rated energy capacity is Power investment cost coefficient of grid-type energy storage system for Energy investment cost coefficient of grid-type energy storage system for The operation and maintenance cost coefficient of grid-type energy storage systems for The charging and discharging efficiencies of the grid-type energy storage system are both 0.95. Among the economic parameters, the discount rate... Pick The lifespan of grid-type energy storage systems The period is 30 years; the penalty coefficient for curtailment of renewable energy. for Network loss cost coefficient for In the frequency safety parameters, the generator's inertia time constant is taken as... The equivalent inertia time constant of the grid-type energy storage system is taken as Maximum allowable frequency variation rate of islanded microgrids for Maximum allowable frequency deviation for islanded microgrids for The method of this invention uses the commercial Gurobi solver to solve the proposed optimization problem, and can also use existing commercial solvers such as CPLEX and MOSEK.

[0328] To compare the impact of different configuration methods on the operational results of isolated microgrids, this embodiment sets up two comparative scenarios. Scenario 1 represents an isolated microgrid using a grid-based energy storage system configuration and scheduling under disturbance conditions; Scenario 2 represents an isolated microgrid using the grid-based energy storage system configuration and scheduling of this invention under disturbance conditions. To cover various seasons, weather conditions, and electricity consumption scenarios, and to improve the generalization ability of the method under different climate and load characteristics, k-means clustering was used to cluster one year's wind power data, solar power data, and load data, resulting in wind power data (wind power), solar power data (solar power), and load data (load power) for four typical days (each typical day is one day). The one year's wind power data, solar power data, and load data were obtained from the Open Data Sets of the IEEE PES Intelligent Systems Subcommittee. Wind power data (WindBased Generation), solar power data (PV Generation), and load data (Consumption) were selected from these sets, and combinations and adjustments can be made according to actual needs. Figure 4 The data includes wind and solar load data for four typical days, among which... Figure 4 (a) in the figure represents the wind power output on four typical days. Figure 4 (b) in the figure represents the photoelectric power on four typical days. Figure 4(c) represents the load power for four typical days. Both Scenario 1 and Scenario 2 are based on this data for optimization.

[0329] The access nodes for the grid-connected energy storage system are nodes 5 and 11. In scenario 1, the battery module configuration results for the grid-connected energy storage system are as follows: on typical day 1, typical day 2, and typical day 4, nodes 5 and 11 are configured respectively. Typical day 3, node 5 configuration Node 11 configuration , This corresponds to one battery module. In Scenario 2, because the grid-type energy storage system needs to simultaneously meet normal scheduling, inertia support, and primary frequency regulation support, the required number of battery modules increases significantly. Under typical day 1 and typical day 3, node 5 is configured with... Node 11 configuration On a typical day 2, both node 5 and node 11 are configured Typical day 4, node 5 configuration Node 11 configuration The comparison shows that, considering frequency safety capacity margin, grid-based energy storage systems can more rationally configure the required number of battery modules compared to grid-connected energy storage systems, thus better ensuring the frequency stability and overall security of isolated microgrids.

[0330] Figure 5 The frequency change rate and frequency deviation for scenario 1 are presented. Due to the lack of consideration for inertia support and the capacity margin of primary frequency regulation, frequency safety limits are exceeded in every typical day. Figure 5 In (a) of the data, the frequency variation rate of the islanded microgrid exceeds [a certain value] over multiple time periods. Limits, of which, during certain periods on typical day 1 and typical day 4, are close to During two typical days, the period is Typical days also have more than 3 The situation. Figure 5 In (b) of the data, the frequency deviation of the islanded microgrid exceeded the limit multiple times. Approaching during certain periods This indicates that, under disturbances, it is difficult to guarantee the frequency security of isolated microgrids without considering the frequency security margin of this invention.

[0331] Figure 6 The rate of frequency change and frequency deviation are given for scenario 2. Figure 5 Compared to (a) in the middle, Figure 6 In (a), the frequency variation rate of the islanded microgrid on four typical days is limited to [missing information]. Within. Of these, typical day 1 and typical day 2 mostly remain within... Typical day 3 is about On a typical day, the four periods of time are close to However, none of them exceeded the limit. Figure 6 In (b) of the above, the overall frequency deviation is lower than that of the previous values. lower than in most periods Approaching in a few periods of time This indicates that the method of the present invention can effectively improve the rate of frequency change and frequency deviation after disturbance.

[0332] Figure 7 The optimal gap (i.e. the distance from the theoretical optimal solution) for different typical days in scenario 2 is given. Figure 7 In (a) of this invention, using the method of this invention (before ablation by the accelerated solution module), the initial optimal gap for some typical days is approximately and in about It later dropped to a lower level (below 1%). Figure 7 In (b) of this invention, after removing the accelerated solution module (after ablation) using the method of this invention, the initial optimal gap for some typical days is close to... It then gradually declined, and at approximately The solution then tends to stabilize. This result demonstrates that the candidate battery module configuration combination screening and convex relaxation processing proposed in this invention significantly improves the solution efficiency.

[0333] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0334] The embodiments described above are merely specific and detailed examples of the embodiments described in this application, and should not be construed as limiting the scope of the application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these modifications and improvements all fall within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the appended claims.

Claims

1. A method for configuring and scheduling a grid-type energy storage system based on frequency security capacity margin, characterized in that: Includes the following steps: S1. Obtain the network topology of the isolated microgrid; S2. Establish a frequency security capacity margin model, and based on the network topology and the frequency security capacity margin model, obtain the dispatchable power and dispatchable energy of each grid-type energy storage system in the islanded microgrid. Then, for each time period... Next node The dispatchable power of a grid-type energy storage system is the node The rated power of the grid-type energy storage system minus the time period Next node The sum of the inertia-supported power and primary frequency regulation-supported power of the grid-type energy storage system, and the dispatchable energy of the grid-type energy storage system at node i during time period t are the values ​​of the node's inertia-supported power and primary frequency regulation-supported power. The rated energy capacity of the grid-type energy storage system minus the time period Next node The reserved energy capacity margin of the grid-type energy storage system , , This is a set of hourly scheduling time periods. Represents the set of nodes in a grid-type energy storage system, time period Next node The inertia support power, primary frequency regulation support power, and reserved energy capacity margin of the grid-type energy storage system are all obtained from the frequency safety capacity margin model. S3. Establish an optimal dispatch model based on the dispatchable power and dispatchable energy of each grid-type energy storage system in the islanded microgrid, where: 1) Objective function: Minimize the sum of investment cost, operation and maintenance cost, and operation and dispatch cost of all grid-type energy storage systems in the isolated microgrid; 2) Constraints: These include configuration constraints, inertia support constraints, frequency security constraints, scheduling constraints, power balance constraints, unit power constraints, voltage constraints, and AC power flow constraints. S4. Use convex relaxation to simplify and optimize the AC power flow constraints in the scheduling model. S5. Solve the optimized scheduling model after convex relaxation based on binary selection variables and commercial solvers to obtain the configuration and scheduling results of the grid-type energy storage system.

2. The configuration and scheduling method for a grid-type energy storage system based on frequency security capacity margin as described in claim 1, characterized in that: The frequency security capacity margin model is constructed as follows: S21. Obtain the disturbance frequency response model of the isolated microgrid; S22. Calculate the time period based on the disturbance frequency response model. Inertia support power of all grid-based energy storage systems in an isolated microgrid ; S23. Calculate the time period based on the disturbance frequency response model. Primary frequency regulation support power of all grid-type energy storage systems in the isolated microgrid ; S24, According to time period A frequency safety capacity margin model is established for the inertia-supported power and primary frequency regulation-supported power of all grid-type energy storage systems in the isolated microgrid. The formula for the frequency safety capacity margin model is as follows: in, Indicates time period Reserved power capacity margin for all grid-type energy storage systems in isolated microgrids; Indicates time period Reserved energy capacity margin for all grid-type energy storage systems in isolated microgrids; Indicates the duration of inertial support, in hours; Indicates the start time of a frequency modulation, in hours; This indicates the time when a single frequency modulation reaches its rated support power, expressed in hours. Indicates the end time of a frequency modulation session, in hours; This indicates the discharge efficiency of a grid-type energy storage system.

3. The configuration and scheduling method for a grid-type energy storage system based on frequency security capacity margin as described in claim 2, characterized in that: The dispatchable power and dispatchable energy of each grid-type energy storage system in the isolated microgrid also satisfy the following conditions: 1) Time period The sum of the inertia support power of all nodes in the grid-type energy storage system is equal to the time period. Inertia support power of all grid-based energy storage systems in an isolated microgrid, time period The sum of the primary frequency regulation support power of all nodes in the grid-type energy storage system is equal to the time period. Primary frequency regulation support power of all grid-type energy storage systems in an isolated microgrid; 2) Time period The sum of the reserved energy capacity margins of all nodes in the grid-type energy storage system equals the time period. Reserved energy capacity margin for all grid-type energy storage systems in the isolated microgrid.

4. The configuration and scheduling method for a grid-type energy storage system based on frequency security capacity margin as described in claim 2, characterized in that: The time period calculated based on the disturbance frequency response model Primary frequency regulation support power of all grid-type energy storage systems in the isolated microgrid The details are as follows: S231, Calculation Period The frequency deviation below is the time period. Active power disturbance of isolated microgrid minus time period Divide the primary frequency regulation support power of all grid-connected energy storage systems in the isolated microgrid by the time period. The load damping coefficient under the condition; S232. Establish frequency deviation constraints, i.e., time period The frequency deviation is less than or equal to the maximum allowable frequency deviation of the islanded microgrid. S233. Substitute the frequency deviation constraint into the formula for the frequency deviation under the corresponding time period to obtain the time period. Primary frequency regulation support power of all grid-type energy storage systems in an isolated microgrid.

5. The configuration and scheduling method for a grid-type energy storage system based on frequency security capacity margin as described in claim 1, characterized in that: The scheduling constraints are as follows: 1) Time period Next node The energy capacity of a grid-type energy storage system is equal to the time period. Next node The energy capacity of the grid-type energy storage system plus its charging efficiency and time period Next node The product of the charging power of the grid-type energy storage system and the time interval between adjacent time periods, minus the reciprocal of the discharge efficiency of the grid-type energy storage system, and the time period. Next node The product of the discharge power of the grid-type energy storage system and the time interval between adjacent time periods; and also greater than or equal to the time interval. Next node The reserved energy capacity margin of the grid-type energy storage system is less than or equal to that of the node. The rated energy capacity of a grid-type energy storage system; 2) Time period Next node The charging power of the grid-type energy storage system is greater than or equal to 0 and less than or equal to the time period. Next node The charge and discharge status of grid-type energy storage systems The product of the rated power of the grid-type energy storage system, And it is greater than or equal to 0 and less than or equal to the time period. Next node The minimum of the maximum charging power and dispatchable power of the grid-type energy storage system; 3) Time period Next node The discharge power of the grid-type energy storage system is greater than or equal to 0 and less than or equal to The product of the rated power of the grid-type energy storage system and the rated power of the grid-type energy storage system; and also greater than or equal to 0 and less than or equal to the time period. Next node The minimum of the following three values ​​for a grid-type energy storage system: the ratio of dispatchable energy to the time interval between adjacent periods, the maximum discharge power, and the dispatchable power.

6. The configuration and scheduling method for a grid-type energy storage system based on frequency security capacity margin as described in claim 1, characterized in that: The simplification of AC power flow constraints in the optimization scheduling model using convex relaxation is as follows: S41, Time Period Acquisition The complex voltage vector of the nodes below; S42, according to time period The construction period of the node complex voltage vector Voltage outer product matrix under; S43, Time-based Solving the voltage outer product matrix under the following time period Next node The active and reactive power injections are calculated, and a second-order cone constraint is introduced.

7. The configuration and scheduling method for a grid-type energy storage system based on frequency security capacity margin as described in claim 6, characterized in that: The second-order cone constraint simultaneously satisfies the following condition: 1) Time period Next node With nodes The L2 norm of the voltage cross product multiplied by 2 is less than or equal to the time interval. Next node With nodes The voltage cross product and time period Next node With nodes The sum of the voltage cross products between them; 2) Time period Next node With nodes The L2 norm of the imaginary part of the voltage cross product between the two time intervals is less than or equal to the time interval. Next node With nodes The voltage cross product and time period Next node With nodes The sum of the voltage cross products between them; 3) Time period Next node With nodes The L2 norm of the voltage cross product between time periods Next node With nodes The L2 norm of the difference in the voltage cross products between the time intervals is less than or equal to the time interval. Next node With nodes The voltage cross product and time period Next node With nodes The sum of the voltage cross products between them.

8. The configuration and scheduling method for a grid-type energy storage system based on frequency security capacity margin as described in claim 1, characterized in that: The optimal scheduling model after convex relaxation is solved based on binary selection variables and a commercial solver to obtain the configuration and scheduling results of the grid-type energy storage system, as detailed below: S51. Exhaustively search for candidate battery module configuration combinations to form a set of candidate battery module configuration combinations, wherein the first... Nodes in the candidate battery module configuration combination The number of battery modules in a grid-type energy storage system , , Nodes in order The lower and upper limits of the number of battery modules in a grid-type energy storage system. , A set of nodes in a network topology. , Indicates the number of candidate battery module configuration combinations; S52. Traverse the candidate battery module configuration combination set, that is, each time select a candidate battery module configuration combination from the candidate battery module configuration combination set based on the binary selection variable. S53. Based on the selected candidate battery module configuration combination, use a commercial solver to solve the optimized scheduling model after convex relaxation processing, and calculate the time period. The inertia support power, primary frequency regulation support power, reserved energy capacity margin, charging power and discharging power of the grid-type energy storage system at each node are used as scheduling parameters. S54. The candidate battery module configuration combination with the objective function of the optimized scheduling model after convex relaxation is minimized is taken as the final battery module configuration combination, and together with the scheduling parameters, it is used as the configuration and scheduling result of the grid-type energy storage system for the corresponding time period.

9. The configuration and scheduling method for a grid-type energy storage system based on frequency security capacity margin as described in claim 8, characterized in that: The time period The inertia support power, primary frequency regulation support power, and reserved energy capacity margin of the grid-type energy storage system at each node are calculated as follows: S531, Calculate the first Nodes in the candidate battery module configuration combination Support allocation coefficient of grid-type energy storage system That is, the first Nodes in the candidate battery module configuration combination The number of battery modules in the grid-type energy storage system and the first The ratio of the sum of the number of battery modules in all nodes of the grid-type energy storage system in a candidate battery module configuration combination; S532, according to the Nodes in the candidate battery module configuration combination Support allocation coefficient of grid-type energy storage system And corresponding binary selection variables, calculate time period Next node Inertia support power of grid-type energy storage systems Time period Next node Primary frequency regulation support power of grid-type energy storage system and time period Next node Reserved energy capacity margin of grid-type energy storage systems The formula is as follows; in, For time period The inertia support power of all grid-type energy storage systems in an isolated microgrid. For time period The primary frequency regulation support power of all grid-type energy storage systems in an isolated microgrid. Indicates time period Reserved energy capacity margin for all grid-type energy storage systems in isolated microgrids Indicates the first Binary selection variables for a combination of candidate battery module configurations.

10. The configuration and scheduling method for a grid-type energy storage system based on frequency security capacity margin as described in claim 1, characterized in that: The commercial solver is one of the following: Gurobi Commercial Solver, CPLEX Commercial Solver, or MOSEK Commercial Solver.

Images