Battery energy storage grid frequency oscillation attack suppression method, system, device and medium based on eigenvalue sensitivity analysis

Abstract

The present application relates to the technical field of power system security, and discloses a battery energy storage power grid frequency oscillation attack suppression method, system, device and medium based on eigenvalue sensitivity analysis. The method comprises: receiving a target power network topology, establishing a first dynamic model representing frequency response and power regulation; establishing a first spatial model for expressing the relationship between each state variable; establishing a first attack model to express the influence of attack disturbance on system matrix and eigenvalue; calculating the eigenvalue change under each attack; calculating the adjustment amount of the energy storage system control parameter on the eigenvalue according to the eigenvalue change; once the adjustment amount restores the eigenvalue to stability, an optimization model is established, and the output result is output, wherein the result is the minimum energy storage adjustment amount. The present application establishes a quantitative matching relationship between attack disturbance and eigenvalue change, guides energy storage parameter adjustment through eigenvalue sensitivity analysis, realizes fast system stability recovery after attack, and reduces adjustment cost.

Description

Technical Field

[0001] This invention relates to the field of power system security technology, specifically to a method, system, device, and medium for suppressing grid frequency oscillation attacks based on eigenvalue sensitivity analysis for battery energy storage. Background Technology

[0002] Power system load frequency control relies on communication networks to transmit signals. Dynamic load disturbance attacks (DLAA) and network parameter attacks (NPA) can both cause system eigenvalues ​​to enter unstable regions, leading to frequency oscillations or even system collapse. Current methods for assessing power system stability under oscillation attacks have several shortcomings: existing research mainly focuses on attack detection and identification, lacking a mechanism for proactively restoring system stability after an attack; a quantitative mapping relationship between attack disturbances and changes in system eigenvalues ​​has not been established, making it impossible to quantify the impact of attacks on stability; and there is no eigenvalue sensitivity analysis, which calculates the changes in eigenvalues ​​after an attack but cannot quantitatively express the adjustment capability of energy storage system control parameters to eigenvalues, failing to provide a sensitivity ranking basis for energy storage parameter adjustment. This results in each energy storage parameter needing to be adjusted individually, leading to low adjustment efficiency and difficulty in supporting the decision-making requirements of rapid stability recovery and minimum cost control. Summary of the Invention

[0003] In view of the above-mentioned existing problems, the present invention provides a method, system, device and medium for suppressing frequency oscillation attacks in battery energy storage power grids based on eigenvalue sensitivity analysis.

[0004] Therefore, the technical problem solved by this invention is: how to establish a quantitative mapping relationship between attack disturbances and eigenvalue changes, and to use eigenvalue sensitivity analysis to guide the adjustment of energy storage parameters, so as to achieve rapid recovery of system stability and minimize adjustment costs after an attack.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for suppressing frequency oscillation attacks in battery energy storage power grids based on eigenvalue sensitivity analysis, comprising, Obtain the topology of the target power network and construct a first dynamic model, which is used to characterize the frequency response and power regulation characteristics of the system. Based on the first dynamic model, a first spatial model is constructed, which is used to characterize the dynamic relationship between the state variables of the system. Based on the first spatial model, a first attack model is constructed. The first attack model is used to quantify the impact of attack disturbances on the system matrix and eigenvalues. In response to the establishment of the first attack model, the change in eigenvalues ​​under each attack is calculated; Based on the change in the characteristic value, calculate the adjustment of the energy storage system control parameters on the characteristic value; In response to the adjustment amount of the eigenvalue, the eigenvalue is restored to stability, an optimization model is constructed, and the output minimizes the energy storage adjustment amount.

[0006] As a preferred embodiment of the battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis described in this invention, the step of obtaining the topology of the target power network and constructing a first dynamic model includes: Obtain the set of generating busbars and load busbars, establish the DC power flow equations, and construct dynamic models and load models for generators and energy storage systems.

[0007] As a preferred embodiment of the battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis described in this invention, wherein: the construction of the first spatial model includes, Based on the dynamic model and load model of the generator and energy storage system, a system state space model is constructed, a system state vector is defined, and the system state matrix and input matrix are obtained. Transform it into a second-order dynamic system model, define the system state variable vector, and obtain the coefficient matrix of the second-order derivative term, the first-order derivative term, and the state variable term.

[0008] As a preferred embodiment of the battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis described in this invention, wherein: the construction of the first attack model includes, By changing the frequency-sensitive load parameters, a dynamic load disturbance attack model is constructed to obtain the system matrix change relationship; By tampering with power transmission lines, a network parameter attack model is constructed to obtain submatrices related to the admittance matrix in the system.

[0009] As a preferred embodiment of the battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis described in this invention, wherein: the calculation of the change in eigenvalues ​​under each attack includes, Based on a second-order dynamic system model, eigenvalue equations are established. Calculate the eigenvalue changes of frequency-sensitive load parameters caused by dynamic load disturbance attacks; Calculate the change in characteristic values ​​of the susceptance parameter of the transmission line caused by the network parameter attack.

[0010] As a preferred embodiment of the battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis described in this invention, wherein: the calculation of the adjustment amount of the energy storage system control parameters to the eigenvalue includes, Based on the change in the characteristic value, the adjustment amount of the characteristic value on the virtual inertia parameter of the energy storage system is calculated. The adjustment amount is used to characterize the energy storage system's ability to adjust the dynamic modes of the system.

[0011] As a preferred embodiment of the battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis described in this invention, the method involves: responding to the adjustment amount of the eigenvalue to restore the eigenvalue to stability, constructing an optimization model, and outputting the minimum energy storage adjustment amount, including: Based on the dynamic load disturbance attack model and the network parameter attack model, energy storage regulation is introduced to obtain the regulated system characteristic values. In response to the adjusted system characteristic values ​​satisfying the constraints, the optimization model is constructed, and the minimum energy storage adjustment amount of the energy storage virtual inertia parameter is output.

[0012] This invention establishes a quantitative relationship between attack disturbances and eigenvalue changes by constructing dynamic models, spatial models, and attack models. This supports eigenvalue sensitivity analysis and energy storage parameter adjustment, enabling rapid recovery of system stability after an attack.

[0013] This invention provides a battery energy storage grid frequency oscillation attack suppression system based on eigenvalue sensitivity analysis, comprising: The dynamic modeling module is used to obtain the topology of the target power network and construct a first dynamic model, which is used to characterize the frequency response and power regulation characteristics of the system. The spatial modeling module is used to construct a first spatial model based on the first dynamic model. The first spatial model is used to characterize the dynamic relationship between various state variables of the system. An attack modeling module is used to construct a first attack model based on the first spatial model. The first attack model is used to quantify the impact of attack perturbations on the system matrix and eigenvalues. The feature value calculation module is used to calculate the change in feature values ​​under each attack in response to the establishment of the first attack model. The adjustment calculation module is used to calculate the adjustment amount of the energy storage system control parameters on the characteristic value based on the change in the characteristic value; An optimized output module is used to restore the stability of the characteristic value in response to the adjustment amount of the characteristic value, construct an optimized model, and output the minimum energy storage adjustment amount.

[0014] The present invention provides a computer device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of a battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis.

[0015] The present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of a method for suppressing frequency oscillation attacks in a battery energy storage grid based on eigenvalue sensitivity analysis are implemented.

[0016] Compared with existing technologies, the beneficial effects of this invention are as follows: First, a dynamic model of the power system is constructed, followed by a state-space and second-order dynamic system model, establishing a quantitative relationship between attack disturbances and eigenvalue changes. The sensitivity of eigenvalues ​​to energy storage control parameters is calculated, and the virtual inertia parameter of the energy storage is adjusted to shift the eigenvalues ​​towards the left half of the complex plane, thereby restoring system stability. Finally, optimization is performed to minimize energy storage regulation while ensuring system stability. Existing technologies mainly focus on attack detection, lacking attention to how to restore system stability after an attack. This invention not only achieves system stability but also clearly identifies the parameters that need adjustment and their adjustment amounts, eliminating the need for trial and error. The most effective energy storage parameters can be determined by calculating sensitivity, thereby improving regulation efficiency. Combined with the optimization model, regulation costs can also be reduced. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0018] Figure 1 This is an implementation diagram of a battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis, provided in one embodiment of the present invention.

[0019] Figure 2 The flowchart illustrates a method for suppressing frequency oscillation attacks in a battery energy storage grid based on eigenvalue sensitivity analysis, as provided in one embodiment of the present invention.

[0020] Figure 3 The generator bus frequency diagram under the attack-free state of the battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis provided in an embodiment of the present invention.

[0021] Figure 4 The generator bus frequency diagram under DLAA attack is provided as an embodiment of the present invention for a battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis.

[0022] Figure 5 The figure shows the distribution of system eigenvalues ​​under a DLAA attack for a battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis, provided in an embodiment of the present invention.

[0023] Figure 6 The generator bus frequency diagram after mitigating DLAA using the Battery Energy Storage Grid Frequency Oscillation Attack Suppression Method based on Eigenvalue Sensitivity Analysis (BESS) is provided as an embodiment of the present invention.

[0024] Figure 7 The figure shows the distribution of system eigenvalues ​​after mitigating DLAA using the BESS method for suppressing frequency oscillation attacks in a battery energy storage grid based on eigenvalue sensitivity analysis, as provided in an embodiment of the present invention.

[0025] Figure 8 The generator bus frequency diagram under NPA attack is provided as an embodiment of the present invention for a battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis.

[0026] Figure 9 The figure shows the distribution of system eigenvalues ​​under NPA attack for a battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis, provided in an embodiment of the present invention.

[0027] Figure 10 The generator bus frequency diagram after mitigating NPA using a battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis, provided as an embodiment of the present invention. Detailed Implementation

[0028] To make the above-mentioned objects, features, and advantages of the present invention more readily understood, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0029] Example 1, referring to Figure 1 The first embodiment of the present invention provides a method for suppressing frequency oscillation attacks in battery energy storage power grids based on eigenvalue sensitivity analysis, comprising: S1: Obtain the topology of the target power network and construct the first dynamic model to characterize the system's frequency response and power regulation characteristics.

[0030] S2: Based on the first dynamic model, construct the first spatial model to characterize the dynamic relationship between the various state variables of the system.

[0031] S3: Based on the first space model, construct the first attack model to quantify the impact of attack perturbations on the system matrix and eigenvalues.

[0032] S4: In response to the establishment of the first attack model, calculate the change in eigenvalues ​​under each attack.

[0033] S5: Calculate the adjustment of the energy storage system control parameters on the characteristic value based on the change in characteristic value.

[0034] S6: In response to the adjustment amount of the eigenvalue, the eigenvalue is restored to stability, an optimization model is constructed, and the output minimizes the energy storage adjustment amount.

[0035] It should be noted that existing technologies mainly focus on attack detection and identification. After an attack occurs, the system becomes unstable but there are no active recovery methods. There is a lack of quantitative mapping between attack disturbances and eigenvalue changes. How to adjust energy storage parameters and by how much depends entirely on experience, resulting in low adjustment efficiency and high cost. Therefore, it is also very important to establish a quantitative relationship between attack disturbances and eigenvalue changes, and to use eigenvalue sensitivity to guide energy storage parameter adjustment, thereby achieving rapid frequency stabilization and minimum cost control.

[0036] This embodiment first establishes a quantitative mapping between a dynamic model and a spatial model, then calculates the sensitivity of eigenvalues ​​to energy storage parameters, and finally uses an optimization model to minimize the adjustment amount while ensuring stability, thereby achieving efficient control from the occurrence of an attack to stable recovery.

[0037] Example 2, refer to Figure 1 As an embodiment of the present invention, based on the above embodiment, a method for suppressing frequency oscillation attacks in battery energy storage power grids based on eigenvalue sensitivity analysis is provided.

[0038] In this embodiment of the application, step S1 involves obtaining the topology of the target power network and constructing a first dynamic model to characterize the system's frequency response and power regulation characteristics. This specifically includes the following steps A1-A3: like Figure 2 As shown, the first dynamic model is a power system dynamic model, and the following is the construction process.

[0039] A1: Obtain the set of generating busbars and the set of load busbars.

[0040] It should be noted that obtaining the topology of the target power network includes the set of all generating buses. and load busbar assembly (The sizes of the sets are respectively) and , The set of all transmission lines is Each transmission line (from the busbar) The starting point and the ending point are the busbars. ).

[0041] A2: Establish DC power flow equations.

[0042] It should be noted that the DC power flow equations are constructed (simplified derivation): in, For the first Net power output of each generator bus For the first Power load of each load bus The voltage phase angle of the generator bus. The voltage phase angle of the load bus. It is an inductor.

[0043] Furthermore, a power flow model containing matrices and vectors is constructed: in, It is the net power output vector of the generating bus (the active power injected into the grid by each generating bus). This is the load power vector of the load bus (the active power consumed by each load bus). The phase angle vector of the generator bus voltage. The phase angle vector of the load bus voltage; matrix , , and Derived from the line parameters This is the matrix formed between the generating buses and the self-admittance term of the generating buses. This is the coupling matrix between the generating bus and the load bus. This is the matrix formed between load buses and the self-admittance terms of the load buses. This is the coupling matrix between the load bus and the generator bus.

[0044] For generator bus : = Generator bus phase angle term + Load bus phase angle term. Stacking up the equations for all generator buses gives us... ; For the load bus By stacking up the equations for all the load buses, we get... .

[0045] A3: Construct dynamic models and load models for generators and energy storage systems.

[0046] It should be noted that a dynamic model of the generator and energy storage system is constructed: in, For the first The rate of change of each phase angle with respect to time For frequency deviation, For system inertia ( ), Damping ratio ( ), This refers to the mechanical power regulation of the generator. For the first Net power output of each generator bus Power BESS.

[0047] in, Virtual inertia adjustment parameters for energy storage systems ( (This is used to describe the inertial support capability of an energy storage system to frequency variations.) The dynamic response time constant of the energy storage system ( ), for The rate of change over time, i.e., the rate of change of BESS injected power.

[0048] According to the secondary frequency control strategy, the generator mechanical power regulation is expressed as: in, and Control coefficient ( ), This refers to the mechanical power regulation of the generator. This is the generator proportional control coefficient. This represents the integral control coefficient for the generator.

[0049] Further, load modeling: Depending on the type of load, the load can be divided into two parts: frequency-sensitive load and constant load.

[0050] No. Each load bus can be represented as: in, The coefficient for frequency-sensitive loads ( (), used to characterize the load's response to frequency changes, This is a quantity related to frequency changes. This is a constant load component.

[0051] Therefore, the load of the entire power system can be expressed as: in, This is the load power vector of the load bus (the active power consumed by each load bus). This is a frequency-sensitive load factor matrix, where the diagonal elements are the frequency-sensitive load factors of each load bus. , For a constant load vector, Let be a frequency change vector, satisfying: This is the phase angle vector of the load bus voltage. This is the rate of change of the phase angle of the load bus voltage, which can also be understood as a quantity related to the frequency deviation.

[0052] Based on step S1, this embodiment establishes dynamic models of generators, energy storage, and loads, as well as DC power flow equations, to characterize the system's frequency response and power regulation characteristics. This lays the foundation for analyzing the impact of the intrusion on the system matrix after the attack and for deriving the stability criteria.

[0053] In this embodiment of the invention, step S2 involves constructing a first spatial model based on the first dynamic model to characterize the dynamic relationships between various state variables of the system, specifically including the following steps B1-B2: The first space model includes the system state space model and the second-order dynamic system model. The construction process is as follows.

[0054] B1: Based on the dynamic model and load model of the generator and energy storage system, construct the system state space model, define the system state vector, and obtain the system state matrix and input matrix.

[0055] It should be noted that the system state-space model is constructed as follows: Based on the aforementioned dynamic and load models of the generator and energy storage system, state-space modeling of the system is performed. Let... ,in, For the first Virtual inertia adjustment parameters for an energy storage system For the first The dynamic response coefficient of an energy storage system and It is an intermediate variable obtained by normalizing the parameters of the energy storage system.

[0056] The system dynamic model is as follows: in, For the frequency deviation vector ( ), The generator inertia coefficient matrix ( ), The generator damping coefficient matrix ( ), The proportional control coefficient matrix in secondary frequency modulation ( ), The integral control coefficient matrix in secondary frequency modulation ( ), For the reason The energy storage equivalent parameter matrix is ​​formed ( ), For the reason The energy storage equivalent parameter matrix is ​​formed ( ), For the virtual inertia adjustment parameter matrix of the energy storage system ( ), The dynamic response coefficient matrix of the energy storage system ( ), The rate of change of the phase angle of the generator bus voltage. The rate of change of phase angle of the load bus voltage. The rate of change of frequency deviation The rate of change matrix for power injection into the energy storage system. For an identity matrix of appropriate size, This is a frequency-sensitive load factor matrix, where the diagonal elements are the frequency-sensitive load factors of each load bus. (Used to describe the degree of response of load power to frequency changes) For constant load vectors.

[0057] Furthermore, the system state vector is defined as follows: Based on the aforementioned model relationships, the system dynamic equations can be rearranged into state-space form: in, The system state matrix, The input matrix is ​​denoted as .

[0058] B2: Transform into a second-order dynamic system model, define the system state variable vector, and obtain the coefficient matrix of the second-order derivative term, the first-order derivative term, and the state variable term.

[0059] It should be noted that the second-order dynamic system model is constructed as follows: To further describe the dynamic characteristics of the system, the state-space model is transformed into a second-order dynamic system model.

[0060] Based on the aforementioned model relationships, and combined with variable relationships By rearranging the system equations, the frequency dynamic equation can be obtained as follows: Meanwhile, the dynamic equation on the load side can be expressed as: Furthermore, according to the dynamic model of the energy storage system, we have: in, For elements is diagonal matrix, For elements is A diagonal matrix. The second rate of change of the frequency deviation. This represents the second-order rate of change of the power injection into the energy storage system. Let be the second rate of change of the phase angle of the load bus voltage. for The first derivative with respect to time is the rate of change of the frequency-related vector.

[0061] Define the state variable vector as follows: The system can then be uniformly represented as the following second-order dynamic form: in, Let be the state vector of the second-order dynamic system. for The first derivative with respect to time, for The second derivative with respect to time, The coefficient matrix of the second derivative terms. Let be the coefficient matrix of the first derivative terms. This is the coefficient matrix of the state variable terms.

[0062] The above second-order dynamic model can describe the dynamic response characteristics of the power system under the influence of energy storage regulation and attack disturbances, providing a basis for subsequent stability analysis.

[0063] Based on step S2, this embodiment establishes a state space and a second-order dynamic model, transforming the dynamic equations of state variables such as frequency, phase angle, and energy storage power into matrix equations. This facilitates quantitative analysis of the impact of the intrusion on the system matrix, calculation of system eigenvalue sensitivity, and derivation of stability criteria after an attack.

[0064] In this embodiment of the application, step S3, based on the first spatial model, constructs a first attack model to quantify the impact of attack perturbations on the system matrix and eigenvalues, specifically including the following steps C1-C2: The first attack model includes the Dynamic Load Disturbance Attack (DLAA) model and the Network Parameter Attack (NPA) model.

[0065] C1: By changing the frequency-sensitive load parameters, a dynamic load disturbance attack model is constructed to obtain the system matrix change relationship.

[0066] It should be noted that the DLAA is achieved by changing the frequency-sensitive load parameters, and its load model is expressed as follows: in, It is a diagonal matrix with diagonal elements. ( busbar Protected; busbar (Vulnerability exists) This represents the load power vector after the Dynamic Load Disturbance Attack (DLAA) is applied.

[0067] Under the attack, the system load characteristics change, leading to a change in the system state-space model. Correspondingly, the system state matrix changes from... The change is as follows: in, The system state matrix under dynamic load disturbance attack. This is the frequency-sensitive load coefficient matrix after a DLAA attack.

[0068] As the system matrix changes, its eigenvalues ​​also change. If eigenvalues ​​exist... satisfy: The system then enters an unstable state, producing frequency oscillations.

[0069] Therefore, by establishing The relationship between dynamic load disturbance attacks and changes in system eigenvalues ​​can be used to analyze the impact of dynamic load disturbance attacks on system stability.

[0070] C2: By tampering with power transmission lines, a network parameter attack model is constructed to obtain the submatrices related to the admittance matrix in the system.

[0071] It should be noted that network parameter attacks are achieved by tampering with the parameters of transmission lines, mainly targeting series compensation devices, thereby changing the line susceptance value and thus affecting the dynamic characteristics of the power system.

[0072] Let the susceptance parameter of the transmission line in the system be... (Connecting busbar) and busbar The susceptance parameters of the transmission line , (Representing the two endpoints of a transmission line), its value changes under attack, which can be expressed as: This causes a change in the system admittance matrix, namely: in, These are the susceptance parameters of the transmission line after NPA application. This is the system admittance matrix of the transmission line after NPA application.

[0073] Furthermore, the submatrices related to the admittance matrix in the system all change accordingly, which can be expressed as: in, This is a submatrix related to the admittance matrix after NPA application. For the set of transmission lines under attack, when At that time, the corresponding line parameters change.

[0074] Under the network parameter attack, the system state-space model changes accordingly, and the corresponding system matrix changes from... The change is as follows: in, This is the system state matrix under a network parameter attack, where the admittance parameter-related parts of the original system matrix are replaced with the parameters after the attack. What was obtained.

[0075] Furthermore, as the system matrix changes, its dynamic modes also change. Let the system eigenvalues ​​before the attack be... The characteristic value after the attack is Then we have: As the system matrix changes, its eigenvalues ​​also change. If there exist eigenvalues ​​that satisfy: The system then enters an unstable state, producing frequency oscillations.

[0076] Therefore, by establishing line parameter disturbances The relationship between the system's characteristic values ​​and the changes in these values ​​can be used to analyze the impact of network parameter attacks on system stability.

[0077] Based on step S3, this embodiment establishes a mathematical model for dynamic load disturbance attack and network parameter attack, characterizing the impact of attack disturbance on the system matrix and the real part of eigenvalues. It can determine whether the attack causes system instability and facilitate the use of energy storage to suppress oscillations after the attack.

[0078] In this embodiment of the application, step S4, in response to the establishment of the first attack model, calculates the change in eigenvalues ​​under each attack, specifically including the following steps D1-D3: D1: Based on the second-order dynamic system model, establish the eigenvalue equations.

[0079] It should be noted that the stability of the system is analyzed based on the above second-order dynamic system model. The second-order dynamic system satisfies the following eigenvalue equations: in, To and The corresponding right eigenvalue, To and The corresponding left eigenvalue.

[0080] D2: Calculate the change in characteristic values ​​of frequency-sensitive load parameters caused by dynamic load disturbance attacks.

[0081] It should be noted that, because the attack targets frequency-sensitive load parameters... Therefore, the system matrix Consequently, changes occur. Based on eigenvalue sensitivity theory, the eigenvalues ​​change with respect to the perturbation parameters. The rate of change can be expressed as: matrix , With parameters Irrelevant ( ),but Consider the set of vulnerable load nodes ,but The change in eigenvalues ​​can then be expressed as: D3: Calculate the change in characteristic value of the transmission line susceptance parameter caused by the network parameter attack.

[0082] It should be noted that network parameter attacks affect the susceptance parameters of transmission lines. Therefore, the part of the system matrix related to the admittance parameter changes. Let the parameter perturbation introduced by the attack be... Under this perturbation, the system matrix changes accordingly, thus affecting the system eigenvalues. Based on a second-order dynamic system model, the eigenvalues ​​change with respect to parameter perturbations. The rate of change can be expressed as: Similarly, it can be simplified to: Furthermore, under the influence of network parameter attacks, the change in system eigenvalues ​​can be expressed as: When eigenvalues ​​exist satisfy: The system then begins to oscillate.

[0083] Based on step S4, this embodiment quantitatively obtains the change in eigenvalues ​​after an attack through eigenvalue sensitivity analysis and establishes an instability criterion that the real part of the eigenvalue is greater than zero, which facilitates the determination of whether the system experiences frequency oscillations after an attack and subsequent energy storage control.

[0084] In this embodiment of the application, step S5, which calculates the adjustment of the energy storage system control parameters on the characteristic value based on the change in characteristic value, specifically includes the following steps: Based on the change in the characteristic value, the adjustment amount of the characteristic value on the virtual inertia parameter of the energy storage system is calculated. The adjustment amount is used to characterize the energy storage system's ability to adjust the dynamic modes of the system.

[0085] It should be noted that, based on the above eigenvalue analysis, the influence of energy storage system control parameters on system eigenvalues ​​is further analyzed.

[0086] in, For the first Virtual inertia parameters of an energy storage system The adjustment amount, For energy storage systems, For the first An energy storage system.

[0087] The above expression can be used to quantitatively describe the energy storage system's ability to regulate the system's dynamic modes.

[0088] Based on step S5, this embodiment calculates the sensitivity of the system characteristic value to the virtual inertia parameter of energy storage, quantifies the influence of energy storage regulation on dynamic modes, and provides directional guidance and control basis for subsequent targeted adjustment of energy storage parameters to suppress frequency oscillations.

[0089] In this embodiment of the application, step S6, which responds to the adjustment amount of the eigenvalue to restore the eigenvalue to stability, constructs an optimization model, and outputs the minimum energy storage adjustment amount, specifically includes the following steps E1-E2: E1: Based on the dynamic load disturbance attack model and the network parameter attack model, energy storage regulation is introduced to obtain the regulated system characteristic values.

[0090] It should be noted that the virtual inertia parameter of the energy storage is adjusted based on the sensitivity results to achieve system stability; Under the influence of Dynamic Load Disturbance Attack (DLAA), energy storage regulation is introduced. The system characteristic value can be expressed as: in, For a bus set with BESS that can defend against DLAA, For the first Frequency-sensitive load factor of each load bus, For DLAA to the first The disturbance introduced by the frequency-sensitive load factor of each load bus.

[0091] Under the influence of Network Parameter Attack (NPA), energy storage regulation is introduced. The first stage of regulation under the combined effect of NPA and BESS is... The system characteristic value can be expressed as: in, This refers to the set of transmission lines affected by the NPA.

[0092] When the following conditions are met: The system then returns to a stable state.

[0093] E2: In response to the adjusted system characteristic values ​​satisfying the constraints, construct the optimization model and output the minimum energy storage adjustment amount of the energy storage virtual inertia parameter.

[0094] It should be noted that, in order to reduce the energy storage regulation cost, the following optimization model is constructed: in, Virtual inertia parameters for energy storage systems Disturbance on the surface for The zero norm is used to count The number of non-zero elements in the middle.

[0095] Furthermore, the constraints are as follows: By solving the above optimization problem, the energy storage regulation can be minimized while ensuring system stability.

[0096] Based on step S6, this embodiment uses eigenvalue sensitivity to guide the adjustment of the virtual inertia parameter of energy storage, so that the real part of the unstable eigenvalue after the attack is adjusted to a negative value, thereby suppressing frequency oscillation; on this basis, a zero-norm optimization model is constructed to optimize the energy storage adjustment amount to reduce control costs without sacrificing system stability.

[0097] Example 3 is the third embodiment of the present invention, which differs from the previous two embodiments in that: This embodiment also provides a battery energy storage grid frequency oscillation attack suppression system based on eigenvalue sensitivity analysis, including: The dynamic modeling module is used to obtain the topology of the target power network and construct a first dynamic model, which is used to characterize the frequency response and power regulation characteristics of the system. The spatial modeling module is used to construct a first spatial model based on the first dynamic model. The first spatial model is used to characterize the dynamic relationship between the various state variables of the system. An attack modeling module is used to construct a first attack model based on the first spatial model. The first attack model is used to quantify the impact of attack perturbations on the system matrix and eigenvalues. The feature value calculation module is used to calculate the change in feature values ​​under each attack in response to the establishment of the first attack model. The adjustment amount calculation module is used to calculate the adjustment amount of the energy storage system control parameters on the characteristic value based on the change in the characteristic value; An optimized output module is used to restore the stability of the characteristic value in response to the adjustment amount of the characteristic value, construct an optimized model, and output the minimum energy storage adjustment amount.

[0098] This embodiment also provides an electronic device applicable to the suppression of frequency oscillation attacks in battery energy storage power grids based on eigenvalue sensitivity analysis, comprising: 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 implement the battery energy storage power grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis proposed in the above embodiment.

[0099] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis proposed in the above embodiments.

[0100] The storage medium proposed in this embodiment and the method for suppressing frequency oscillation attacks in battery energy storage grids based on eigenvalue sensitivity analysis proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0101] Example 4, the fourth embodiment of the present invention, provides simulation verification of a battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis, including: The method of the present invention is verified using the IEEE 9-bus power system as an example.

[0102] Table 1 Simulation parameters of the IEEE 9-node system

[0103] in, , , The generator's inertia coefficient. , , This is the generator damping coefficient. , , This is the proportional control coefficient. , , This is the integral control coefficient. For frequency-sensitive load factor, These are the susceptance parameters of the transmission line (where the subscripts indicate the busbar numbers at both ends). , , The dynamic response coefficient of BESS. , , These are the virtual inertia adjustment parameters for BESS.

[0104] As shown in Table 1, a dynamic model of the power system is constructed based on the system topology and line parameters, and a corresponding state-space model and second-order dynamic system model are established.

[0105] Under normal operating conditions, all system eigenvalues ​​are located in the left half of the complex plane, and the system is in a stable state.

[0106] DLAA and its mitigation measures: DLAA attack parameters are set to... .

[0107] It can be observed that the frequencies of buses 7, 8, and 9 in the IEEE 9-node system oscillate and deviate from the standard values, indicating the existence of system eigenvalues ​​with non-negative real parts. DLAA attacks are effective against the IEEE 9-node power system. Figure 3 As shown.

[0108] Furthermore, using the method of this invention, BESS parameters are designed to alleviate DLAA. The BESS parameters are then adjusted. Relief performance such as Figure 4 As shown.

[0109] The above results indicate that BESS can mitigate the impact of DLAA on grid frequency.

[0110] NPA and its mitigation measures: NPA attack parameters are set to... , .

[0111] It can be observed that the frequencies of buses 7, 8, and 9 in the IEEE 9-node system oscillate and deviate from the standard values, indicating the existence of system eigenvalues ​​with non-negative real parts. NPA attacks are effective against the IEEE 9-node power system. Figure 5 As shown.

[0112] Furthermore, using the method of this invention, BESS parameters are designed to alleviate NPA. The BESS parameters are then adjusted. Relief performance such as Figure 6 As shown.

[0113] The above results indicate that BESS can mitigate the impact of NPA on grid frequency.

[0114] The method of the present invention is further verified using the IEEE 39-bus power system as an example.

[0115] BESS parameters: , .

[0116] Frequency-sensitive load factor: Under normal circumstances, all generator bus frequencies converge to the rated value (60Hz), and the system does not exhibit any unstable dynamic modes.

[0117] Construct the corresponding dynamic model and second-order dynamic system model based on the system parameters.

[0118] Then, following the steps described in S4, analyze the sensitivity of the system eigenvalues ​​to the DLAA, NPA, and BESS parameters, and calculate the eigenvalue sensitivity accordingly: in, For the system number The sensitivity index of the eigenvalue to dynamic load disturbance attack parameters is used to measure the sensitivity of the eigenvalue to the dynamic load disturbance attack parameters. Changes in the frequency-sensitive load factor of the individual load bus The larger the value, the more likely the load bus is to cause system modal changes after being attacked by DLAA, and the more obvious the impact on system stability. For the system number A sensitivity index of eigenvalues ​​to network parameter attack parameters, used to measure the susceptance disturbance of transmission lines. The degree of impact on system stability; the larger the value, the easier it is for the system characteristic value to change after the corresponding transmission line parameters are tampered with, and the more sensitive the line is to NPA. For the system number The sensitivity index of each eigenvalue to the BESS control parameters is used to measure the adjustment of the virtual inertia parameter of the energy storage system. The ability to adjust system eigenvalues; the larger the value, the better. The stronger a BESS's ability to regulate system modes, the more suitable it is as a control object for suppressing oscillation attacks.

[0119] By using eigenvalue sensitivity mapping for oscillation attacks, defenders can rank the vulnerabilities of load buses and transmission lines to DLAA and NPA separately. Based on the eigenvalue sensitivity mapping of BESS parameters, defenders can design the most effective mitigation strategies.

[0120] DLAA and its mitigation measures: DLAA attack parameters are set to... .

[0121] It can be observed that the generator frequencies at IEEE 39-node oscillate and deviate from the standard values, indicating the existence of system eigenvalues ​​with non-negative real parts. DLAA attacks are effective against the IEEE 39-node power system. Figure 7 As shown.

[0122] Furthermore, using the method of this invention, BESS parameters are designed to alleviate DLAA. The BESS parameters are then adjusted. Relief performance such as Figure 8 As shown.

[0123] Therefore, in larger power systems, BESS can be used to suppress DLAA.

[0124] Furthermore, consider the minimum-cost DLAA from the attacker's perspective. An attacker can disrupt a single vulnerable load bus, thereby triggering an oscillation.

[0125] Furthermore, analyzing the lowest-cost mitigation strategy reveals that only the BESS parameters of the IEEE 39-node power system need to be changed. This can reduce the impact of minimum cost DLAA.

[0126] NPA and its mitigation measures: NPA attack parameters are set to... .

[0127] It can be observed that the frequencies of buses 32, 36, and 39 at IEEE 39-node oscillate and deviate from the standard values, indicating the existence of system eigenvalues ​​with non-negative real parts. NPA attacks are effective against the IEEE 39-node power system. Figure 9 As shown.

[0128] Furthermore, using the method of this invention, BESS parameters are designed to alleviate NPA. The BESS parameters are then adjusted. Relief performance such as Figure 10 As shown.

[0129] Therefore, in larger power systems, BESS can be used to suppress NPA.

[0130] Furthermore, consider the minimum-cost NPA from the attacker's perspective. An attacker can cause power system frequency oscillations by modifying the inductance on only one line.

[0131] Furthermore, analysis of the lowest-cost mitigation strategy reveals that at least two BESS parameters are required to reduce the impact of the lowest-cost NPA.

[0132] Comparative analysis of the system frequency response results before, after, and after the attack shows that the method of the present invention can effectively suppress oscillation attacks and improve the system's operational stability in power systems of different scales.

[0133] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0134] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for suppressing frequency oscillation attacks in battery energy storage power grids based on eigenvalue sensitivity analysis, characterized in that: include, Obtain the topology of the target power network and construct a first dynamic model, which is used to characterize the frequency response and power regulation characteristics of the system. Based on the first dynamic model, a first spatial model is constructed, which is used to characterize the dynamic relationship between the state variables of the system. Based on the first spatial model, a first attack model is constructed. The first attack model is used to quantify the impact of attack disturbances on the system matrix and eigenvalues. In response to the establishment of the first attack model, the change in eigenvalues ​​under each attack is calculated; Based on the change in the characteristic value, calculate the adjustment of the energy storage system control parameters on the characteristic value; In response to the adjustment amount of the eigenvalue, the eigenvalue is restored to stability, an optimization model is constructed, and the output minimizes the energy storage adjustment amount.

2. The method for suppressing frequency oscillation attacks in battery energy storage power grids based on eigenvalue sensitivity analysis as described in claim 1, characterized in that: The process of obtaining the topology of the target power network and constructing the first dynamic model includes: Obtain the set of generating busbars and load busbars, establish the DC power flow equations, and construct dynamic models and load models for generators and energy storage systems.

3. The method for suppressing frequency oscillation attacks in battery energy storage power grids based on eigenvalue sensitivity analysis as described in claim 2, characterized in that: The construction of the first spatial model includes, Based on the dynamic model and load model of the generator and energy storage system, a system state space model is constructed, a system state vector is defined, and the system state matrix and input matrix are obtained. Transform it into a second-order dynamic system model, define the system state variable vector, and obtain the coefficient matrix of the second-order derivative term, the first-order derivative term, and the state variable term.

4. The method for suppressing frequency oscillation attacks in battery energy storage power grids based on eigenvalue sensitivity analysis as described in claim 3, characterized in that: The construction of the first attack model includes, By changing the frequency-sensitive load parameters, a dynamic load disturbance attack model is constructed to obtain the system matrix change relationship; By tampering with power transmission lines, a network parameter attack model is constructed to obtain submatrices related to the admittance matrix in the system.

5. The method for suppressing frequency oscillation attacks in battery energy storage power grids based on eigenvalue sensitivity analysis as described in claim 4, characterized in that: The calculation of the change in feature values ​​under each attack includes, Based on a second-order dynamic system model, eigenvalue equations are established. Calculate the eigenvalue changes of frequency-sensitive load parameters caused by dynamic load disturbance attacks; Calculate the change in characteristic values ​​of the susceptance parameter of the transmission line caused by the network parameter attack.

6. The method for suppressing frequency oscillation attacks in battery energy storage power grids based on eigenvalue sensitivity analysis as described in claim 5, characterized in that: The adjustment amount of the energy storage system control parameters to the characteristic value includes, Based on the change in the characteristic value, the adjustment amount of the characteristic value on the virtual inertia parameter of the energy storage system is calculated. The adjustment amount is used to characterize the energy storage system's ability to adjust the dynamic modes of the system.

7. The method for suppressing frequency oscillation attacks in battery energy storage power grids based on eigenvalue sensitivity analysis as described in claim 6, characterized in that: In response to the adjustment amount of the eigenvalue, which restores the eigenvalue to stability, an optimization model is constructed, and the output minimizes the energy storage adjustment amount, including: Based on the dynamic load disturbance attack model and the network parameter attack model, energy storage regulation is introduced to obtain the regulated system characteristic values. In response to the adjusted system characteristic values ​​satisfying the constraints, the optimization model is constructed, and the minimum energy storage adjustment amount of the energy storage virtual inertia parameter is output.

8. A battery energy storage grid frequency oscillation attack suppression system based on eigenvalue sensitivity analysis, wherein the battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis as described in any one of claims 1 to 7 is characterized in that, include: The dynamic modeling module is used to obtain the topology of the target power network and construct a first dynamic model, which is used to characterize the frequency response and power regulation characteristics of the system. The spatial modeling module is used to construct a first spatial model based on the first dynamic model. The first spatial model is used to characterize the dynamic relationship between various state variables of the system. An attack modeling module is used to construct a first attack model based on the first spatial model. The first attack model is used to quantify the impact of attack perturbations on the system matrix and eigenvalues. The feature value calculation module is used to calculate the change in feature values ​​under each attack in response to the establishment of the first attack model. The adjustment calculation module is used to calculate the adjustment amount of the energy storage system control parameters on the characteristic value based on the change in the characteristic value; An optimized output module is used to restore the stability of the characteristic value in response to the adjustment amount of the characteristic value, construct an optimized model, and output the minimum energy storage adjustment amount.

9. A computer 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 battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the battery energy storage grid frequency oscillation attack suppression method based on eigenvalue sensitivity analysis as described in any one of claims 1 to 7.

Images