Island micro-grid optimal scheduling method and system considering whole frequency response process

Abstract

The invention discloses an island micro-grid optimization scheduling method and system considering a frequency response whole process, and belongs to the field of island micro-grid optimization scheduling, and the method comprises the steps: generating a photovoltaic output typical day scene through K-means clustering, and constructing an island micro-grid basic random optimization scheduling model; dividing the frequency response of the island micro-grid after disturbance into three parts; constructing an island microgrid disturbance power and mechanical inertia expression based on frequency response characteristics under generator N-1 off-grid disturbance; and aiming at the three parts of frequency response, linear frequency stability constraints are respectively constructed by coordinating multiple frequency modulation resources, and the linear frequency stability constraints are integrated into a basic random optimization scheduling model of the island micro-grid, so that the island micro-grid optimization scheduling method considering the whole process of frequency response is formed. According to the method, the stable frequency under the N-1 event is guaranteed, the source-load-storage response potential is mined, the operation cost and the dependence on the generator are reduced, and safe and efficient operation of the island microgrid is supported.

Description

Technical Field

[0001] This invention belongs to the field of islanded microgrid optimization scheduling, specifically relating to an islanded microgrid optimization scheduling method and system that considers the entire frequency response process. Background Technology

[0002] Guided by global goals for clean energy transition and sustainable development, islanded microgrids, with their flexible integration capabilities of distributed energy resources, have become a core solution for reliable power supply to isolated areas such as islands and ships, playing an increasingly prominent role in improving the cleanliness of energy supply and ensuring energy security. With the rapid development of inverter-based resources such as photovoltaics and energy storage, their penetration rate in islanded microgrids continues to increase, accelerating the transformation of the energy structure towards low carbon emissions, but also bringing new technological challenges. These resources themselves lack the inertia and frequency regulation capabilities of traditional generators, resulting in islanded microgrids generally facing problems of insufficient inertia and a lack of primary frequency regulation capacity. Furthermore, the islanded operation characteristics prevent them from obtaining frequency support from the external power grid, making them prone to large frequency fluctuations when encountering power disturbances, seriously threatening the safe and stable operation of the system.

[0003] Currently, the core contradiction in islanded microgrid dispatch lies in balancing frequency stability assurance with operational economy. Existing frequency control research largely relies on the assumption of sufficient frequency regulation reserves, but pre-reserving these reserves weakens the system's operational economy. Furthermore, different control methods offer varying frequency regulation support effects, and traditional dispatch models often fail to comprehensively cover the entire frequency response process, hindering the full exploitation of the collaborative frequency regulation potential of various resources. How to optimize dispatch strategies to reduce operating costs while ensuring frequency stability has become a critical issue urgently needing resolution in the development of islanded microgrids. Therefore, developing optimized dispatch technologies for islanded microgrids that consider the entire frequency response process, fully leveraging the frequency regulation capabilities of diverse resources such as generators, energy storage, and photovoltaics, and achieving a balance between safety, stability, and operational economy has become a core requirement for supporting the high-quality development of islanded microgrids. Summary of the Invention

[0004] The purpose of this invention is to address the technical problems of existing isolated microgrids, such as insufficient inertia and limited primary frequency regulation capacity due to the increased penetration of inverter-type resources like photovoltaics, leading to significant frequency fluctuations when encountering power disturbances. It also addresses the difficulty in balancing frequency stability and operational economy in existing isolated microgrid scheduling, and the fact that reserving frequency regulation reserves weakens system operational economy. This invention provides an optimized scheduling method that divides the frequency response into three stages and constructs targeted linearized frequency stability constraints. Through a basic stochastic optimization scheduling model containing a two-stage cost objective function, it fully explores the collaborative frequency regulation potential of generators, photovoltaics, energy storage systems, and loads, reducing operating costs and dependence on traditional generators. This ensures frequency stability under N-1 events, supporting the safe and stable operation of isolated microgrids.

[0005] According to one aspect of the present invention, an optimized scheduling method for islanded microgrids considering the entire frequency response process is provided, comprising: K-means clustering was used to generate typical daily photovoltaic power output scenarios; Based on an islanded microgrid containing generators, photovoltaics, energy storage systems and loads, a basic stochastic optimization scheduling model for the islanded microgrid with a two-stage cost objective function and multi-dimensional constraints is constructed. The frequency response of the islanded microgrid after the disturbance power is divided into three parts: the inertial response at the initial moment of the disturbance, the first frequency response after the disturbance occurs, and the second frequency response after the frequency reaches the quasi-steady state. Based on the frequency response characteristics of generator N-1 disconnection disturbance, the disturbance power and mechanical inertia of the islanded microgrid are obtained; Based on the three parts after frequency response division, and combined with the disturbance power and mechanical inertia of the islanded microgrid, linearized frequency stability constraints are constructed respectively, and integrated into the basic stochastic optimization scheduling model of the islanded microgrid to complete the optimization scheduling of the islanded microgrid considering the entire frequency response process.

[0006] Furthermore, a basic stochastic optimization scheduling model for islanded microgrids, containing a two-stage cost objective function and multi-dimensional constraints, is constructed, including: A two-stage cost objective function is constructed, including a first-stage cost objective function and a second-stage cost objective function. The first-stage cost objective function includes the start-up and shutdown cost of the generator, the power generation cost, and the frequency regulation reserve cost of the energy storage system. The second-stage cost objective function includes the primary and secondary reserve costs of the generator, the primary reserve cost of the photovoltaic system, and the cost of unnecessary load shedding. Construct multi-dimensional constraints, including power balance constraints within the islanded microgrid, generator output limit constraints, generator primary and secondary frequency regulation reserve constraints, energy storage system operation constraints, energy storage system frequency regulation reserve constraints, photovoltaic frequency regulation reserve constraints, and non-essential load shedding power constraints.

[0007] A basic stochastic optimization scheduling model for islanded microgrids is constructed based on the stage cost objective function and multi-dimensional constraints.

[0008] Furthermore, the frequency response of the islanded microgrid after disturbance power is divided into three parts, including: Assuming the disturbance occurs t At time 0, in the frequency response, the frequency of the islanded microgrid is... t At time 1, it reaches a quasi-steady state. t The value was restored to its nominal value at time 2. The inertial response at the initial moment after the perturbation is divided t At time 0, the time period corresponding to the first frequency modulation response after the disturbance occurs (t 0, t 1], the time period corresponding to the second frequency response after the frequency reaches quasi-steady state ( t 1, t 2).

[0009] Furthermore, based on the frequency response characteristics of generator N-1 under grid disconnection disturbance, the disturbance power and mechanical inertia of the islanded microgrid are obtained, including: The disturbance power of the isolated microgrid is the instantaneous change in the active power of the shut-down generator, expressed as:

[0010] in, The time delay for removing unnecessary loads; The time step for discretizing the frequency response process; The disturbance power that occurs in event N-1; For the unit Unnecessary load shedding during a network outage event; The first indicates the photovoltaic output A typical scenario; Indicates the first One scheduling period; The power of the disturbance is indicated by the number of seconds after the disturbance. A discrete point; The mechanical inertia of the islanded microgrid is expressed as the sum of the mechanical inertia of all online units except for the generators that are shut down.

[0011] Furthermore, based on the three parts after frequency response partitioning, and considering the disturbance power and mechanical inertia of the islanded microgrid, linearized frequency stability constraints are constructed respectively, including: Based on the inertial response at the initial moment of the disturbance, constraints are imposed on the response power of the energy storage system and the frequency change rate of the isolated microgrid. Power limiting is considered for the inertial response power of the energy storage system, and the corresponding equality constraints are obtained. For the nonlinear elements in the equality constraints, the Big M method is used to convert them into a linear combination of binary variables and continuous variables, and finally the frequency change rate constraints of the isolated microgrid are obtained. Based on the frequency response after the disturbance occurs, constraints are imposed on the frequency deviation after the disturbance, the steady-state frequency deviation, and the response power of each frequency modulation resource. Based on the secondary frequency response after the frequency reaches quasi-steady state, the secondary frequency regulation reserve of the islanded microgrid is constrained; it is assumed that all generators in the grid can participate in secondary frequency regulation, and the recovery of unnecessary load shedding is not considered.

[0012] Furthermore, based on the inertial response at the initial moment of the disturbance, the frequency change rate constraint of the islanded microgrid is obtained, including:

[0013] in, For the maximum allowable rate of frequency change, The inertial response power of the energy storage system at the initial moment of the disturbance after constraint. The disturbance power that occurs in event N-1. Indicates generator Mechanical inertia of islanded microgrids under outage N-1 event Indicates time.

[0014] Furthermore, constraints are imposed on the frequency deviation after disturbance, the steady-state frequency deviation, and the response power of each frequency modulation resource, including: Construct a frequency response model for an islanded microgrid system that considers power limiting; Based on the frequency response model of an islanded microgrid system, the coupling relationship between the response power and frequency deviation of each resource is handled by a discretization method. The power limiting is achieved by applying a penalty variable and constructing a linearized frequency stability constraint.

[0015] Furthermore, following the second frequency response, it also includes: The different N-1 events are distinguished using a secondary frequency modulation reserve, expressed as:

[0016] in, This is the sum of the steady-state frequency response power when all frequency modulation resources are constrained. To run the generator g During the period t For off-grid generators The secondary frequency modulation reserve released at that time, The disturbance power that occurs in event N-1. For the unit Unnecessary load shedding during a network outage event; This is the reference power.

[0017] According to one aspect of the present invention, an optimized scheduling system for islanded microgrids considering the entire frequency response process is provided, comprising: The data processing module is used to classify and aggregate the acquired historical photovoltaic power output time series data using K-means clustering to generate typical daily photovoltaic power output scenarios. The module for constructing a basic stochastic optimization scheduling model for islanded microgrids is used to construct a basic stochastic optimization scheduling model for islanded microgrids containing generators, photovoltaics, energy storage systems and loads, with a two-stage cost objective function and multi-dimensional constraints. The response partitioning module is used to divide the frequency response of the islanded microgrid after a disturbance into three parts: the inertial response at the initial moment of the disturbance, the primary frequency response after the disturbance occurs, and the secondary frequency response after the frequency reaches a quasi-steady state. The frequency response characteristic module is used to obtain the disturbance power and mechanical inertia of the islanded microgrid based on the frequency response characteristics of the generator N-1 disconnection disturbance. The islanded microgrid optimization scheduling module is used to construct linearized frequency stability constraints based on the three parts after frequency response division, combined with the disturbance power and mechanical inertia of the islanded microgrid, and integrate them into the basic stochastic optimization scheduling model of the islanded microgrid to complete the optimization scheduling of the islanded microgrid considering the entire frequency response process.

[0018] According to one aspect of the present invention, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the islanded microgrid optimal scheduling method that considers the entire frequency response process.

[0019] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention uses K-means clustering to generate typical daily photovoltaic power output scenarios and constructs a basic stochastic optimization scheduling model with a two-stage cost objective function. This effectively handles the uncertainty of photovoltaic power output and covers generator start-up and shutdown costs, energy storage frequency regulation backup costs, and multi-resource backup and unnecessary load shedding costs. It takes into account both operational economy and frequency regulation backup requirements, and reduces dependence on traditional generators.

[0020] 2. This invention divides the frequency response of an isolated microgrid after disturbance into three parts, and constructs linearized frequency stability constraints for each part, including the cooperating generator, energy storage, and photovoltaic multi-frequency regulation resources. This precisely solves the problems of insufficient system inertia and lack of primary frequency regulation capacity caused by the increased penetration rate of inverter-type resources, and ensures that the frequency change rate, maximum frequency difference, and steady-state frequency difference are all within safe thresholds under the N-1 event.

[0021] 3. This invention adopts a method for constructing expressions of disturbance power and mechanical inertia based on the N-1 grid disconnection disturbance characteristics of generators. It formulates a dedicated non-essential load shedding strategy for each generator grid disconnection event, avoiding the problem of excessive or insufficient load shedding caused by the traditional approach that only considers the shutdown of the maximum capacity unit. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This invention provides a schematic flowchart of an islanded microgrid optimization scheduling method that considers the entire frequency response process.

[0024] Figure 2 This is a schematic diagram of an islanded microgrid provided by the present invention.

[0025] Figure 3 This is a schematic diagram of the frequency change curve of the entire process after a low-frequency disturbance in an islanded microgrid provided by the present invention.

[0026] Figure 4 This is a schematic diagram of the frequency response model of an islanded microgrid system provided by the present invention.

[0027] Figure 5 Box plots of frequency indices after N-1 events occur in various time periods of the islanded microgrid embodiment provided by the present invention, with and without considering frequency stability constraints. Detailed Implementation

[0028] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] Specifically, such as Figure 1 As shown, this embodiment of the invention proposes an optimized scheduling method for islanded microgrids that considers the entire frequency response process, including: Step 1: Generating typical daily photovoltaic power output scenarios using K-means clustering; Step 2: For islanded microgrids containing generators, photovoltaics, energy storage systems, and loads, such as... Figure 2As shown, a basic stochastic optimization scheduling model for an islanded microgrid with a two-stage cost objective function and multi-dimensional constraints is constructed. This is a stochastic optimization scheduling model for the islanded microgrid without frequency stability constraints, containing an optimal cost objective function representing economic efficiency and constraints representing the operating state. Step 3: The frequency response of the islanded microgrid after a disturbance is divided into three parts: the inertial response at the initial moment of the disturbance, the primary frequency regulation response after the disturbance, and the secondary frequency response after the frequency reaches a quasi-steady state. Step 4: Because the frequency response of the islanded microgrid occurs after the disturbance power, and the inertial response during the frequency response process is inseparable from the mechanical inertia of the islanded microgrid, therefore... Based on the frequency response characteristics of generator N-1 grid disconnection disturbance, this paper constructs expressions for the disturbance power and mechanical inertia of the islanded microgrid to facilitate subsequent linearization modeling of the frequency response process. Step 5: For the three parts after frequency response division, considering the inertial response, primary frequency response, and secondary frequency response of the generator, the inertial response and primary frequency response of the energy storage system, the primary frequency response of the photovoltaic system, and the role of unnecessary load shedding in the primary and secondary frequency responses, linearized frequency stability constraints are constructed respectively, and integrated into the basic stochastic optimization scheduling model of the islanded microgrid to form an islanded microgrid optimization scheduling method that considers the entire frequency response process.

[0030] Specifically, the embodiments of the present invention provide the specific content of step 1: Historical photovoltaic power output time series data for the current month are selected to construct a basic dataset to ensure that the data matches the meteorological characteristics and sunshine patterns of the scheduling period, providing data support for subsequent scenario generation; The K-means clustering algorithm, which is a mature application in the field of power system data dimensionality reduction and feature extraction, is selected to balance the computational efficiency required for scheduling optimization with the representativeness of the scenario. The original photovoltaic power output samples were classified and aggregated using the K-means clustering algorithm to complete data dimensionality reduction and feature extraction, and to select representative sample clusters. Based on the clustering results, generate s A typical photovoltaic power output scenario is used to transform the uncertain photovoltaic power output data in the scheduling into deterministic power output data that can be directly calculated under discrete typical scenarios, laying the foundation for subsequent optimization scheduling modeling.

[0031] Specifically, the embodiments of the present invention provide the following details for step 2: The cost objective function of the basic stochastic optimization scheduling model for islanded microgrids is set as the sum of two-stage cost objective functions. The first-stage cost objective function... The second-stage cost objective function includes generator start-up and shutdown costs, power generation costs, and frequency regulation reserve costs of the energy storage system. This includes the costs of primary and secondary frequency regulation reserves for generators, the primary reserve cost of the photovoltaic system, and the cost of unnecessary load shedding. The mathematical expression for this cost objective function is shown below:

[0032] (1) (2) (3) in: , , , These are collections of scheduling periods, generators, typical photovoltaic scenarios, and off-grid units; Reference power; , , Generators g Start-up costs, shutdown costs, and unit power generation costs; , These are binary variables representing the generator's start-up and shutdown, respectively. This refers to the generator's power output. The scheduling duration is measured in units. The frequency regulation reserve cost coefficient for energy storage systems; For frequency regulation reserves of energy storage systems; The probability of a typical photovoltaic power generation scenario; , , These are the reserve cost coefficients for primary frequency regulation of synchronous machines, secondary frequency regulation of synchronous machines, and primary frequency regulation of photovoltaic systems, respectively. , , These are respectively the primary frequency regulation reserve of the synchronous machine, the secondary frequency regulation reserve of the synchronous machine, and the primary frequency regulation reserve of the photovoltaic system; The unit cost of removing unnecessary loads is considered. Since the frequency regulation reserve variable of the energy storage system is coupled with many first-stage variables, it is treated as a first-stage variable to simplify model complexity and improve computational efficiency.

[0033] The constraints in the basic stochastic optimal scheduling model for islanded microgrids are as follows: (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) in: To contribute to the actual development of photovoltaics; , These are the charging and discharging power of the energy storage system, respectively. The load power within the microgrid; A binary variable representing the start and stop of the generator; , These are the maximum and minimum generating power of generator g, respectively; The secondary frequency regulation reserve release rate of generator g; Allowable generator secondary frequency regulation reserve release time; A binary variable representing the charging and discharging of the energy storage system; a value of 1 indicates that the energy storage system is charging. This represents the maximum charge / discharge rate of the energy storage system. Energy for energy storage systems; , These are the charging and discharging efficiencies of the energy storage system; For frequency regulation energy storage of energy storage systems; MPPT power of photovoltaics; This represents the minimum allowable unloaded ratio for photovoltaic systems. The allowable value for unnecessary load shedding for each scheduling period. Equation (4) is the power balance constraint in the isolated microgrid, ensuring power matching between the generation side and the consumption side; Equation (5) limits the upper and lower technical limits of the generator output in the start-stop state; Equation (6) constrains the total output of the generator set including primary and secondary reserves to not exceed the maximum allowable value, where the primary and secondary reserves are nominal values; Equation (7) limits the range of secondary reserve capacity of the generator set; Equations (8)-(9) limit the charging and discharging power of energy storage within the allowable range; Equation (10) constrains the total power of energy storage including frequency regulation reserves; Equation (11) describes the dynamic change of energy storage energy with charging, discharging and efficiency; Equation (12) ensures the energy balance of energy storage at the beginning and end of the cycle; Equations (13)-(14) respectively limit the lower limit of energy storage energy and the upper limit when including reserves; Equation (15) constrains the actual output of photovoltaic within the range of MPPT value and allowable load reduction; Equation (16) defines the photovoltaic primary reserve capacity as the difference between MPPT output and actual output; Equation (17) limits the maximum value of unnecessary load shedding. These constraints, from the perspectives of power, energy, and backup, ensure the feasibility and stability of islanded microgrid scheduling and operation.

[0034] Specifically, the embodiments of the present invention provide the specific content of step 3: In an isolated microgrid, if any grid-connected synchronous generator unexpectedly disconnects from the grid, it will cause a drop in system frequency. By leveraging the coordinated regulation of various frequency regulation resources within the microgrid, the active power deficit can be gradually offset, restoring the system frequency to its nominal value. Figure 3 Described t The frequency trajectory of the microgrid after the disturbance occurs at time 0. To support the construction of subsequent full-process frequency constraints, this invention divides the frequency response process of the microgrid into the inertial response at the initial moment of the disturbance (corresponding to...). t 0) The first frequency modulation response phase after the disturbance occurs (corresponding time period ( t 0, t 1]) and the secondary frequency response stage after the frequency reaches a quasi-steady state (corresponding time period ( t 1, t 2).

[0035] Specifically, the embodiments of the present invention provide the specific content of step 4: In an isolated microgrid, a sudden outage of a single generator can occur at any time, and the corresponding disturbance power can be described as the instantaneous change in the generator's active power. Let... The disturbance power that occurs in the N-1 event (corresponding to the off-grid generator) ), Let g be the power output of generator g, when g is equal to... When representing the same generator, the following equation can be established:

[0036] (18) If system frequency regulation resources are insufficient, cutting off some unnecessary loads can effectively ensure frequency stability. However, existing literature on N-1 events typically only considers the shutdown scenario of the largest capacity unit. The load shedding strategies developed based on this have the following drawbacks when other units shut down: not shedding loads may lead to frequency instability, while shedding loads may cause excessive frequency overshoot due to over-cutting. Therefore, this paper develops corresponding unnecessary load shedding strategies for each generator's sudden shutdown event to avoid this problem. Furthermore, unnecessary load shedding is affected by measurement and communication processes, and cannot act instantaneously during disturbances, resulting in a certain time delay. Therefore, the disturbance power of the system under N-1 events can be expressed in discrete form as:

[0037] (19) in, The time delay for removing unnecessary loads; The time step for discretizing the frequency response process; The disturbance power that occurs in event N-1; For the unit Unnecessary load shedding during a network outage event; The first indicates the photovoltaic output A typical scenario; Indicates the first One scheduling period; The power of the disturbance is indicated by the number of seconds after the disturbance. A discrete point.

[0038] The mechanical inertia of an islanded microgrid system is provided solely by online and operational generators; therefore, for generators... The mechanical inertia of a microgrid under an N-1 shutdown event can be expressed as: (20) in, Let g be the per-unit value of the inertial time constant of the generator, representing the value of the energy storage system's own inertial time constant converted to the system's reference power. This indicates a generator that is offline, while g indicates a generator that is running.

[0039] Specifically, the embodiments of the present invention provide the specific content of step 5: For the three parts of the frequency response, linearized frequency stability constraints are constructed by coordinating multiple frequency modulation resources, as described below.

[0040] 1) Inertial response at the initial moment of the disturbance At the initial moment of the disturbance, constraints are imposed on the additional output power of the energy storage system and the frequency change rate of the microgrid.

[0041] The frequency response of an islanded microgrid can be expressed by a swing equation in the form of a differential equation: (twenty one) in, The rate of change of frequency in a frequency response process; For system damping; This refers to the frequency deviation during a frequency response process. This refers to the primary frequency regulation response power of the generator; The frequency response power of the energy storage system; This refers to the frequency response power of photovoltaics. This represents the disturbance power.

[0042] After the disturbance power occurs, the energy storage system can provide virtual inertia and virtual damping. If there is no frequency regulation reserve limitation, its unrestricted response power expression can be stated as: (twenty two) in, This represents the per-unit value of the inertia coefficient of the energy storage system. This represents the per-unit value of the damping coefficient for the energy storage system.

[0043] However, due to the frequency regulation reserve constraints of the energy storage system, the actual additional output power of the energy storage system is: (twenty three) At the initial moment of the disturbance, i.e. Time (corresponding) Figure 3 middle t 0), the system frequency deviation and the output power of resources participating in primary frequency regulation are both 0, so the unrestricted response power, the actual response power, and the swing equation of the energy storage system can be expressed as follows: (twenty four) (25) (26) It is clear that due to the presence of the min term, equation (7) exhibits nonlinearity. We will now address this nonlinearity. First, we assume that the inertial response power of the energy storage system is unconstrained at the initial moment. Then, equation (8) can be transformed into:

[0044] (27) Combining equation (6) and equation (9), we can obtain: (28) Thus, we can see that, due to It consists of binary variables that enable and disable the decision-making unit. The product of binary variables and continuous variables can be precisely transformed using the Big M method, thus transforming equation (10) into a linear constraint. Meanwhile, for the three decision variables appearing in equation (7)... Nonlinear constraints can still be transformed using the Big M method. The Big M method replaces the product of a binary variable and a continuous variable with a new variable. Using a sufficiently large positive number M, constraints are constructed such that this new variable is 0 when the binary variable is 0, and exactly equal to the continuous variable when it is 1. This transforms the product of the two into a linear constraint. Substituting the corresponding variables into the model, we obtain the following formula:

[0045] (29) (30) (31) (32) (33) (34) (35) in, For the defined auxiliary variables; The inertial response power of the energy storage system at the initial moment of an unconstrained disturbance; It is a sufficiently large constant; The inertial response power of the energy storage system at the initial moment of the disturbance after constraint; The binary variable is introduced for linearization of the min step. Equation (12) is linearized to obtain equations (13) and (14); Equation (11) is substituted with text variables and linearized to obtain equation (15); Equations (17) and (18) are the linearization transformation formulas of equation (16).

[0046] There exists a maximum rate of frequency change during the frequency response process at the initial moment of the disturbance, which needs to be constrained. The limited additional power of energy storage can be regarded as a reduction in the disturbance power; therefore, the rate of frequency change constraint of the islanded microgrid can be written as:

[0047] (36) in, This represents the maximum allowable rate of frequency change for the system.

[0048] 2) The primary frequency modulation response following the disturbance In the primary frequency regulation phase following a disturbance, the frequency response can be modeled by constructing a primary frequency regulation model and employing a discretization method. The system frequency response model of a microgrid is as follows: Figure 4 As shown in the figure, this stage constrains the frequency deviation, steady-state frequency deviation, and response power of each frequency regulation resource in the microgrid.

[0049] The discretized form of the linearized rocking equation can be written as: (37) (38) (39) (40) in, For the defined auxiliary variables; The frequency deviation at discrete time points after discretization; , , These are the frequency response power of the generator, energy storage system, and photovoltaic system, respectively.

[0050] The dynamic primary frequency regulation response constraints, primary reserve constraints, and primary ramp rate limit constraints of generators and photovoltaics can be addressed by adjusting the following: Figure 4 The Laplace operator is discretized, and modeled using a method analogous to the response modeling of the following energy storage system. The frequency response constraints and storage constraints of the energy storage system are shown below:

[0051] (41) (42) (43) in, This refers to the unconstrained response power of the energy storage system in the subsequent stages of a disturbance. This refers to the response power of the energy storage system that is limited in the subsequent stages of a disturbance. This is an auxiliary variable that acts on the limiting stage of the energy storage system.

[0052] At the steady-state operating point of the frequency response, the frequency change rate of the microgrid is 0. Therefore, the following equation can be derived from the system's swing equation: (44) in, It is the sum of the steady-state frequency response power when all frequency modulation resources are constrained in steady state; This refers to the steady-state frequency deviation. , , These are the steady-state frequency response power of the generator, energy storage system, and photovoltaic system, respectively.

[0053] The primary frequency response constraints and corresponding frequency modulation reserve constraints for the frequency modulation resources at the steady-state operating point are as follows: (45) (46) (47) (48) (49) (50) (51) (52) (53) in, , , These are the steady-state frequency response power of unrestricted generators, photovoltaic systems, and energy storage systems, respectively. , These are the droop coefficients for the generator and the photovoltaic system, respectively. , , These are the steady-state frequency response power of the generator, photovoltaic, and energy storage systems after the restrictions were applied. , , These are auxiliary variables that limit the steady-state frequency response power of generators, photovoltaic systems, and energy storage systems, respectively.

[0054] In the first frequency response stage, the frequency deviation and steady-state frequency deviation of all discrete points in this stage are constrained as follows: (54) (55) in, , These are the system's limits for maximum frequency difference and maximum rate of frequency change, respectively.

[0055] To ensure the mandatory enforcement of power limiting, a penalty cost is added as the cost objective function, which can be expressed as: (56) in, It is a sufficiently small constant; , These are auxiliary variables that act on the generator and photovoltaic limiting components, respectively.

[0056] Therefore, the final cost objective function of this optimization scheduling model can be expressed as: (57) 3) Second frequency response after the frequency reaches quasi-steady state Once the microgrid's frequency response reaches a quasi-steady state, assuming all generators within the grid can participate in the secondary frequency response, and without considering the recovery from unnecessary load shedding, the system frequency can recover from the post-disturbance steady state to the nominal frequency through the secondary frequency response. Meanwhile, since the secondary frequency regulation process does not involve frequency security issues, only constraints on the secondary frequency regulation reserve need to be considered.

[0057] At the end of the second frequency response, the sum of the reserved secondary frequency regulation capacity and the primary frequency regulation capacity should completely compensate for the system power deficit. It should be noted that the secondary frequency regulation reserve released by the same generator is not the same when other generators disconnect from the grid. Therefore, a variable... Differentiating the secondary frequency modulation reserve under different N-1 events, we have the following formula:

[0058] (58) dynamo g During the period t against Secondary frequency modulation reserves released when disconnected from the grid It should not exceed the generator g During the period t generator g Total reserves, this constraint can be written as: (59) Meanwhile, since the energy storage system will continue to compensate for part of the power deficit during the second frequency response phase by generating additional power as required by the frequency quasi-steady state, it is necessary to reserve sufficient energy reserves for the energy storage system during the second frequency response phase. This can be expressed as a constraint: (60) To demonstrate the advantages of this invention in ensuring frequency stability, an islanded microgrid consisting of four conventional synchronous generators, loads, an energy storage power station, and a photovoltaic power station was constructed to analyze the model and method proposed in this invention. The rated frequency of the system is 50Hz, and the frequency change rate, maximum frequency deviation, and steady-state frequency deviation safety threshold are set to ±1.25Hz / s, ±0.8Hz, and ±0.5Hz, respectively. Box plots of the microgrid's frequency performance after the N-1 generator disconnection event at various time periods, considering and not considering frequency stability constraints, are shown below. Figure 5As shown, without considering frequency stability constraints, the frequency distribution of each unit after disconnection from the grid in each time period exceeds the limit; after considering frequency stability constraints, the frequency distribution of each unit after disconnection from the grid in each time period can be guaranteed to be within a safe range.

[0059] The implementation of the various embodiments of the present invention is based on programmed processing by a device with processor functionality. Therefore, in practical engineering, the technical solutions and functions of the various embodiments of the present invention are encapsulated into various modules. Based on this reality, and building upon the above embodiments, the embodiments of the present invention provide an islanded microgrid optimal scheduling system that considers the entire frequency response process. This system is used to execute an islanded microgrid optimal scheduling method that considers the entire frequency response process from the above method embodiments.

[0060] The system includes: a data processing module for generating typical daily photovoltaic power output scenarios using K-means clustering; a basic stochastic optimization scheduling model construction module for islanded microgrids, used to construct a basic stochastic optimization scheduling model for islanded microgrids containing generators, photovoltaics, energy storage systems, and loads, with a two-stage cost objective function and multi-dimensional constraints; a response partitioning module for dividing the frequency response of the islanded microgrid after disturbance power into three parts: the inertia response at the initial moment of the disturbance, the primary frequency response after the disturbance, and the secondary frequency response after the frequency reaches a quasi-steady state; a frequency response characteristic module for obtaining the disturbance power and mechanical inertia of the islanded microgrid based on the frequency response characteristics under generator N-1 grid disconnection disturbance; and an islanded microgrid optimization scheduling module for constructing linearized frequency stability constraints based on the three parts of the frequency response partitioning, combined with the disturbance power and mechanical inertia of the islanded microgrid, and integrating them into the basic stochastic optimization scheduling model for islanded microgrids to complete the optimization scheduling of the islanded microgrid considering the entire frequency response process.

[0061] The islanded microgrid optimization scheduling system provided in this invention addresses the technical problems of existing islanded microgrids, such as insufficient inertia and limited primary frequency regulation capacity due to the increased penetration of inverter-type resources like photovoltaics, leading to significant frequency fluctuations when encountering power disturbances. It also addresses the difficulty in balancing frequency stability and operational economy in existing islanded microgrid scheduling, where reserving frequency regulation reserves weakens system operational economy. The system employs several modules and utilizes a basic stochastic optimization scheduling model with a two-stage cost objective function to fully explore the collaborative frequency regulation potential of generators, photovoltaics, energy storage systems, and loads. This reduces operating costs and dependence on traditional generators, ensuring frequency stability under N-1 events and supporting the safe and stable operation of islanded microgrids.

[0062] Based on the same inventive concept as the foregoing embodiments, this embodiment of the invention also provides an electronic device, including a memory and a processor. The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize an islanded microgrid optimization scheduling method that considers the entire frequency response process as proposed in the above embodiments.

[0063] Finally, it should be noted that the above specific embodiments are merely representative examples of the present invention. Obviously, the present invention is not limited to the above specific embodiments and many variations are possible. Any simple modifications, equivalent changes, and alterations made to the above specific embodiments based on the technical essence of the present invention should be considered within the protection scope of the present invention.

Claims

1. A method for optimal scheduling of islanded microgrids considering the entire frequency response process, characterized in that, include: K-means clustering was used to generate typical daily photovoltaic power output scenarios; Based on an islanded microgrid containing generators, photovoltaics, energy storage systems and loads, a basic stochastic optimization scheduling model for the islanded microgrid with a two-stage cost objective function and multi-dimensional constraints is constructed. The frequency response of the islanded microgrid after the disturbance power is divided into three parts: the inertial response at the initial moment of the disturbance, the first frequency response after the disturbance occurs, and the second frequency response after the frequency reaches the quasi-steady state. Based on the frequency response characteristics of generator N-1 disconnection disturbance, the disturbance power and mechanical inertia of the islanded microgrid are obtained; Based on the three parts after frequency response division, and combined with the disturbance power and mechanical inertia of the islanded microgrid, linearized frequency stability constraints are constructed respectively, and integrated into the basic stochastic optimization scheduling model of the islanded microgrid to complete the optimization scheduling of the islanded microgrid considering the entire frequency response process.

2. The islanded microgrid optimal scheduling method considering the entire frequency response process according to claim 1, characterized in that, A basic stochastic optimization scheduling model for islanded microgrids, containing a two-stage cost objective function and multi-dimensional constraints, is constructed, including: A two-stage cost objective function is constructed, including a first-stage cost objective function and a second-stage cost objective function. The first-stage cost objective function includes the start-up and shutdown cost of the generator, the power generation cost, and the frequency regulation reserve cost of the energy storage system. The second-stage cost objective function includes the primary and secondary reserve costs of the generator, the primary reserve cost of the photovoltaic system, and the cost of unnecessary load shedding. Construct multi-dimensional constraints, including power balance constraints within the islanded microgrid, generator output limit constraints, generator primary and secondary frequency regulation reserve constraints, energy storage system operation constraints, energy storage system frequency regulation reserve constraints, photovoltaic frequency regulation reserve constraints, and non-essential load shedding power constraints. A basic stochastic optimization scheduling model for islanded microgrids is constructed based on the stage cost objective function and multi-dimensional constraints.

3. The islanded microgrid optimal scheduling method considering the entire frequency response process according to claim 1, characterized in that, The frequency response of an islanded microgrid after disturbance power is divided into three parts, including: Assuming the disturbance occurs t At time 0, in the frequency response, the frequency of the islanded microgrid is... t At time 1, it reaches a quasi-steady state. t The value will be restored to the nominal value at time 2. The inertial response at the initial moment after the perturbation is divided t At time 0, the time period corresponding to the first frequency modulation response after the disturbance occurs ( t 0, t 1], the time period corresponding to the second frequency response after the frequency reaches quasi-steady state ( t 1, t 2).

4. The islanded microgrid optimal scheduling method considering the entire frequency response process according to claim 1, characterized in that, Based on the frequency response characteristics of generator N-1 disconnection disturbance, the disturbance power and mechanical inertia of the islanded microgrid are obtained, including: The disturbance power of the isolated microgrid is the instantaneous change in the active power of the shut-down generator, expressed as: ， in, The time delay for removing unnecessary loads; The time step for discretizing the frequency response process; The disturbance power that occurs in event N-1; For the unit Unnecessary load shedding during a network outage event; The first indicates the photovoltaic output A typical scenario; Indicates the first One scheduling period; The power of the disturbance is indicated by the number of seconds after the disturbance. A discrete point; The mechanical inertia of the islanded microgrid is expressed as the sum of the mechanical inertia of all online units except for the generators that are shut down.

5. The islanded microgrid optimal scheduling method considering the entire frequency response process according to claim 1, characterized in that, Based on the three parts of the frequency response partitioning, and considering the disturbance power and mechanical inertia of the islanded microgrid, linearized frequency stability constraints are constructed, including: Based on the inertial response at the initial moment of the disturbance, constraints are imposed on the response power of the energy storage system and the frequency change rate of the isolated microgrid. Power limiting is considered for the inertial response power of the energy storage system, and the corresponding equality constraints are obtained. For the nonlinear elements in the equality constraints, the Big M method is used to convert them into a linear combination of binary variables and continuous variables, and finally the frequency change rate constraints of the isolated microgrid are obtained. Based on the frequency response after the disturbance occurs, constraints are imposed on the frequency deviation after the disturbance, the steady-state frequency deviation, and the response power of each frequency modulation resource. Based on the secondary frequency response after the frequency reaches quasi-steady state, the secondary frequency regulation reserve of the islanded microgrid is constrained; it is assumed that all generators in the grid can participate in secondary frequency regulation, and the recovery of unnecessary load shedding is not considered.

6. The islanded microgrid optimal scheduling method considering the entire frequency response process according to claim 5, characterized in that, Based on the inertial response at the initial moment of the disturbance, the frequency change rate constraint of the islanded microgrid is obtained, including: ， in, For the maximum allowable rate of frequency change, The inertial response power of the energy storage system at the initial moment of the disturbance after constraint. The disturbance power that occurs in event N-1. Indicates generator Mechanical inertia of islanded microgrids under outage N-1 event Indicates time.

7. The islanded microgrid optimal scheduling method considering the entire frequency response process according to claim 5, characterized in that, Constraints are imposed on the frequency deviation after disturbance, the steady-state frequency deviation, and the response power of each frequency modulation resource, including: Construct a frequency response model for an islanded microgrid system that considers power limiting; Based on the frequency response model of an islanded microgrid system, the coupling relationship between the response power and frequency deviation of each resource is handled by a discretization method. The power limiting is achieved by applying a penalty variable and constructing a linearized frequency stability constraint.

8. The islanded microgrid optimal scheduling method considering the entire frequency response process according to claim 5, characterized in that, Following the second frequency response, it also includes: The different N-1 events are distinguished using a secondary frequency modulation reserve, expressed as: ， in, This is the sum of the steady-state frequency response power when all frequency modulation resources are constrained. To run the generator g During the period t For off-grid generators The secondary frequency modulation reserve released at that time, The disturbance power that occurs in event N-1. This is to address unnecessary load shedding during unit disconnection events; This is the reference power.

9. An optimized scheduling system for islanded microgrids considering the entire frequency response process, characterized in that, include: The data processing module is used to classify and aggregate the acquired historical photovoltaic power output time series data using K-means clustering to generate typical daily photovoltaic power output scenarios. The module for constructing a basic stochastic optimization scheduling model for islanded microgrids is used to construct a basic stochastic optimization scheduling model for islanded microgrids containing generators, photovoltaics, energy storage systems and loads, with a two-stage cost objective function and multi-dimensional constraints. The response partitioning module is used to divide the frequency response of the islanded microgrid after a disturbance into three parts: the inertial response at the initial moment of the disturbance, the primary frequency response after the disturbance occurs, and the secondary frequency response after the frequency reaches a quasi-steady state. The frequency response characteristic module is used to obtain the disturbance power and mechanical inertia of the islanded microgrid based on the frequency response characteristics of the generator N-1 disconnection disturbance. The islanded microgrid optimization scheduling module is used to construct linearized frequency stability constraints based on the three parts after frequency response division, combined with the disturbance power and mechanical inertia of the islanded microgrid, and integrate them into the basic stochastic optimization scheduling model of the islanded microgrid to complete the optimization scheduling of the islanded microgrid considering the entire frequency response process.

10. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the islanded microgrid optimal scheduling method that considers the entire frequency response process as described in any one of claims 1 to 8.

Images