Power network-oriented energy storage integrated anti-spoofing attack frequency control method

Abstract

The invention relates to a power network-oriented energy storage integrated anti-spoofing attack frequency control method and device, computer equipment, a computer readable storage medium and a computer program product, which can be used in the technical field of power. The method comprises the following steps: acquiring system operation parameters of a power system comprising an energy storage unit, and constructing a frequency control model according to the system operation parameters; constructing a spoofing attack model aiming at the frequency control model according to the frequency-driven spoofing attack information, and constructing a closed-loop control system model comprising the frequency control model and the spoofing attack model according to the PI control information; performing stability analysis processing on the closed-loop control system model, and determining a constraint condition enabling the closed-loop control system model to meet a preset robust performance index; determining gain information of a target PI controller according to the constraint condition; and performing frequency control on the power system according to the gain information of the target PI controller. By adopting the method, the frequency control stability of the power system can be improved.

Description

Technical Field

[0001] This application relates to the field of power technology, and in particular to a frequency control method, apparatus, computer equipment, computer-readable storage medium, and computer program product for power grids that integrates energy storage and resists spoofing attacks. Background Technology

[0002] In the field of power system control technology, frequency control is a key link in ensuring the safe and stable operation of the power grid. However, with the increasing networking and intelligence of the power system, the power system is facing increasingly severe cybersecurity threats.

[0003] Existing frequency control schemes are mainly designed for physical faults and lack the ability to defend against network spoofing attacks, resulting in low frequency control stability of power systems. Summary of the Invention

[0004] Therefore, it is necessary to provide a frequency control method, device, computer equipment, computer-readable storage medium, and computer program product for power networks that integrates energy storage and resists deception attacks, which can improve the frequency control stability of power systems and address the aforementioned technical problems.

[0005] Firstly, this application provides a frequency control method for integrated energy storage and anti-spoofing attacks in power networks. The method includes:

[0006] Obtain the system operating parameters of the power system containing energy storage units, and construct a frequency control model containing dynamic characteristic information of generator sets and dynamic characteristic information of energy storage systems based on the system operating parameters;

[0007] Based on preset frequency-driven deception attack information, a deception attack model targeting the frequency control model is constructed, and based on preset PI control information, a closed-loop control system model including the frequency control model and the deception attack model is constructed.

[0008] The closed-loop control system model is subjected to stability analysis to determine the constraints that enable the closed-loop control system model to meet the preset robust performance index.

[0009] Based on the constraints, determine the target PI controller gain information;

[0010] Frequency control of the power system is performed based on the target PI controller gain information.

[0011] In one embodiment, constructing a deception attack model targeting the frequency control model based on preset frequency-driven deception attack information includes:

[0012] Based on the frequency-driven spoofing attack information, determine the attack weight parameters;

[0013] Based on the attack weight parameters, the frequency information in the frequency control model is subjected to attack perturbation processing to obtain damaged frequency information;

[0014] Based on the attack weight parameters, the regional control error signal in the frequency control model is subjected to attack perturbation processing to obtain the damaged regional control error signal;

[0015] The deception attack model is constructed based on the damaged frequency information and the damaged area control error signal.

[0016] In one embodiment, the stability analysis of the closed-loop control system model includes:

[0017] Construct a game theory model based on controller and attacker information;

[0018] Based on the game theory model, a target performance index is determined; the target performance index is used to characterize the controller's ability to suppress disturbance signals under the most unfavorable attack conditions.

[0019] Based on the target performance indicators, the closed-loop control system model is subjected to stability analysis.

[0020] In one embodiment, determining the target PI controller gain information based on the constraints includes:

[0021] Within the preset gain parameter range, initialize the candidate gain parameter population;

[0022] Based on the constraints, fitness identification processing is performed on each candidate gain parameter in the candidate gain parameter population to obtain the fitness value of each candidate gain parameter.

[0023] Based on the fitness values ​​of each candidate gain parameter, the candidate gain parameter population is iteratively updated to obtain the target gain parameter;

[0024] The target PI controller gain information is determined based on the target gain parameter.

[0025] In one embodiment, constructing a frequency control model that includes dynamic characteristic information of the generator set and dynamic characteristic information of the energy storage system based on the system operating parameters includes:

[0026] Based on the system operating parameters, determine the power balance information of the power system;

[0027] Based on the power balance information, state space information is determined; the state variables in the state space information include frequency deviation, governor valve position deviation, turbine mechanical power deviation, and energy storage unit output power deviation.

[0028] The frequency control model is constructed based on the state space information.

[0029] In one embodiment, the closed-loop control system model includes composite disturbance information;

[0030] The method further includes:

[0031] Obtain external disturbance information and determine the attack disturbance information based on the deception attack model;

[0032] The external disturbance information and the attack disturbance information are fused to obtain the composite disturbance information.

[0033] Secondly, this application also provides a frequency control device for power grids that integrates energy storage and resists spoofing attacks. The device includes:

[0034] The parameter acquisition module is used to acquire the system operating parameters of the power system containing the energy storage unit, and to construct a frequency control model containing the dynamic characteristic information of the generator set and the dynamic characteristic information of the energy storage system based on the system operating parameters.

[0035] The model building module is used to build a deception attack model against the frequency control model based on preset frequency-driven deception attack information, and to build a closed-loop control system model containing the frequency control model and the deception attack model based on preset PI control information.

[0036] The model analysis module is used to perform stability analysis on the closed-loop control system model and determine the constraints that make the closed-loop control system model meet the preset robust performance index.

[0037] The information determination module is used to determine the target PI controller gain information based on the constraints.

[0038] The system control module is used to perform frequency control on the power system based on the target PI controller gain information.

[0039] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0040] Obtain the system operating parameters of the power system containing energy storage units, and construct a frequency control model containing dynamic characteristic information of generator sets and dynamic characteristic information of energy storage systems based on the system operating parameters;

[0041] Based on preset frequency-driven deception attack information, a deception attack model targeting the frequency control model is constructed, and based on preset PI control information, a closed-loop control system model including the frequency control model and the deception attack model is constructed.

[0042] The closed-loop control system model is subjected to stability analysis to determine the constraints that enable the closed-loop control system model to meet the preset robust performance index.

[0043] Based on the constraints, determine the target PI controller gain information;

[0044] Frequency control of the power system is performed based on the target PI controller gain information.

[0045] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, performs the following steps:

[0046] Obtain the system operating parameters of the power system containing energy storage units, and construct a frequency control model containing dynamic characteristic information of generator sets and dynamic characteristic information of energy storage systems based on the system operating parameters;

[0047] Based on preset frequency-driven deception attack information, a deception attack model targeting the frequency control model is constructed, and based on preset PI control information, a closed-loop control system model including the frequency control model and the deception attack model is constructed.

[0048] The closed-loop control system model is subjected to stability analysis to determine the constraints that enable the closed-loop control system model to meet the preset robust performance index.

[0049] Based on the constraints, determine the target PI controller gain information;

[0050] Frequency control of the power system is performed based on the target PI controller gain information.

[0051] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, performs the following steps:

[0052] Obtain the system operating parameters of the power system containing energy storage units, and construct a frequency control model containing dynamic characteristic information of generator sets and dynamic characteristic information of energy storage systems based on the system operating parameters;

[0053] Based on preset frequency-driven deception attack information, a deception attack model targeting the frequency control model is constructed, and based on preset PI control information, a closed-loop control system model including the frequency control model and the deception attack model is constructed.

[0054] The closed-loop control system model is subjected to stability analysis to determine the constraints that enable the closed-loop control system model to meet the preset robust performance index.

[0055] Based on the constraints, determine the target PI controller gain information;

[0056] Frequency control of the power system is performed based on the target PI controller gain information.

[0057] The aforementioned frequency control method, apparatus, computer equipment, computer-readable storage medium, and computer program product for power grids with integrated energy storage and anti-spoofing attack capabilities acquire system operating parameters of a power system including energy storage units. Based on these system operating parameters, a frequency control model incorporating dynamic characteristic information of generator sets and energy storage systems is constructed. Based on preset frequency-driven spoofing attack information, a spoofing attack model targeting the frequency control model is constructed. Based on preset PI control information, a closed-loop control system model incorporating both the frequency control model and the spoofing attack model is constructed. Stability analysis is performed on the closed-loop control system model to determine constraints that ensure the model meets preset robust performance indicators. Based on these constraints, target PI controller gain information is determined. Frequency control of the power system is then performed based on the target PI controller gain information. This scheme acquires the system operating parameters of the power system including energy storage units and constructs a frequency control model that includes dynamic characteristic information of generator sets and energy storage systems, which is beneficial for comprehensively characterizing the dynamic characteristics of the power system. By constructing a deception attack model targeting the frequency control model and establishing a closed-loop control system model, it is beneficial for incorporating attack behavior into the control system analysis framework. By performing stability analysis on the closed-loop control system model and determining the constraints that meet the preset robust performance indicators, it is beneficial for ensuring the stability of the system under attack conditions. By determining the target proportional-integral controller gain information based on the constraints and performing frequency control, it is beneficial for the power system to maintain efficient and robust frequency regulation performance even when subjected to deception attacks, thereby improving the frequency control stability of the power system. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 This is a flowchart illustrating a frequency control method for integrated energy storage and anti-spoofing attacks for power networks in one embodiment.

[0060] Figure 2 This is a schematic diagram of a power system load frequency control topology containing an energy storage unit in one embodiment;

[0061] Figure 3 This is a schematic diagram of frequency deviation under different control strategies in one embodiment;

[0062] Figure 4 This is a structural block diagram of an integrated energy storage and anti-spoofing frequency control device for power grids in one embodiment.

[0063] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0065] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0066] In one exemplary embodiment, such as Figure 1As shown, a frequency control method for integrated energy storage to resist spoofing attacks in power networks is provided. This embodiment illustrates the method by applying it to a terminal. It is understood that this method can also be applied to a server, or to a system including both a terminal and a server, and is implemented through interaction between the terminal and the server. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, etc.; the server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. In this embodiment, the method includes the following steps:

[0067] Step S101: Obtain the system operating parameters of the power system containing the energy storage unit, and construct a frequency control model containing the dynamic characteristic information of the generator set and the dynamic characteristic information of the energy storage system based on the system operating parameters.

[0068] Step S102: Based on the preset frequency-driven deception attack information, construct a deception attack model targeting the frequency control model, and based on the preset PI control information, construct a closed-loop control system model that includes the frequency control model and the deception attack model.

[0069] Step S103: Perform stability analysis on the closed-loop control system model to determine the constraints that enable the closed-loop control system model to meet the preset robust performance index.

[0070] Step S104: Determine the target PI controller gain information based on the constraints.

[0071] Step S105: Perform frequency control on the power system based on the target PI controller gain information.

[0072] The power system can be a load frequency control system that includes generator sets and energy storage units. The power system achieves dynamic frequency regulation through power balance equations.

[0073] Among them, the energy storage unit can be a supercapacitor energy storage device. The energy storage unit is used to assist the generator set in frequency regulation and improve the flexibility of system frequency regulation.

[0074] Among them, the system operating parameters can be the dynamic characteristic parameters of the power system, including the equivalent inertia constant, damping coefficient, turbine time constant, governor time constant, droop coefficient, energy storage unit time constant, and unit adjustment factor.

[0075] Among them, the frequency control model can be a load frequency control dynamic model based on state-space equations. The frequency control model regards the main generator, governor, turbine and energy storage system as a unified dynamic object, and establishes the dynamic relationship of load frequency control through power balance equations.

[0076] Among them, the dynamic characteristic information of the generator set can be dynamic characteristic description information including parameters such as the time constant of the governor and turbine, and the droop coefficient.

[0077] Among them, the dynamic characteristic information of the energy storage system can be dynamic characteristic description information including parameters such as the time constant, unit adjustment factor, and participation ratio of the energy storage unit.

[0078] Among them, frequency-driven spoofing attack information can be fake data injection attack information based on frequency signals. Attackers dynamically disturb frequency measurement values ​​by forging or delaying the transmission of control data.

[0079] Among them, the deception attack model can be a fake data injection attack signal model that introduces a time-varying attack weight matrix. The deception attack model is used to describe the attacker's tampering behavior with frequency signals and regional control error signals.

[0080] Among them, proportional-integral control information (PI control information) can be the control strategy information of the proportional-integral controller, including proportional gain parameters and integral gain parameters.

[0081] Among them, the closed-loop control system model can be a system state-space model that integrates the frequency control model, the deception attack model, and the proportional-integral controller.

[0082] Among them, the stability analysis process can be based on the robust stability analysis method of Lyapunov-Krasovsky function, which verifies the asymptotic stability of the system by constructing a functional.

[0083] Among them, the preset robust performance index can be a disturbance suppression level constraint condition. The preset robust performance index is used to measure the system's performance under the most unfavorable attack conditions.

[0084] Among them, the constraint condition can be a stability condition that satisfies the linear matrix inequality. The constraint condition unifies the stability constraint and the disturbance suppression performance index in the same analysis framework.

[0085] The target proportional-integral controller gain information can be the optimal proportional gain parameter and integral gain parameter obtained through optimization algorithm search. The target proportional-integral controller gain information enables the closed-loop control system model to satisfy the linear matrix inequality stability condition and performance constraints.

[0086] Frequency control can be a control process that adjusts the frequency deviation of the power system based on the gain information of the target proportional-integral controller.

[0087] Optionally, the terminal acquires system operating parameters of the power system including energy storage units. These parameters include equivalent inertia constant, damping coefficient, turbine time constant, governor time constant, droop coefficient, energy storage unit time constant, and unit adjustment factor. Based on these parameters, the terminal constructs a frequency control model incorporating dynamic characteristic information of both the generator set and the energy storage system. This model treats the main generator, governor, turbine, and energy storage system as a unified dynamic object, establishing a dynamic relationship for load frequency control through power balance equations. The terminal then constructs a deception attack model targeting the frequency control model based on preset frequency-driven deception attack information. The model introduces a time-varying attack weight matrix to model the tampering of frequency signals and regional control error signals. The terminal constructs a closed-loop control system model, including a frequency control model and a spoofing attack model, based on preset proportional-integral (PI) control information. The terminal performs stability analysis on the closed-loop control system model using the Lyapunov-Krasovsky function to determine constraints that ensure the model meets preset robust performance indicators. These constraints are stability conditions that satisfy linear matrix inequalities. Based on these constraints, the terminal uses an optimization algorithm to search for and determine the target PI controller gain information. Finally, the terminal performs frequency control on the power system based on the target PI controller gain information.

[0088] For example, the terminal acquires system operating parameters of a power system including energy storage units. These parameters include equivalent inertial constant, damping coefficient, turbine time constant, governor time constant, droop coefficient, energy storage unit time constant, unit adjustment factor, governor participation ratio, and energy storage system participation ratio. Based on these parameters, the terminal establishes a state-space equation model and constructs a frequency control model incorporating generator dynamic characteristic information and energy storage system dynamic characteristic information. This model considers time-varying transmission delays caused by communication networks and pulse disturbances caused by sudden external interference. Based on preset frequency-driven deception attack information, the terminal constructs a deception attack model targeting the frequency control model. This model introduces a time-varying attack weight matrix to describe scaling and disturbance attacks and establishes a tampering model for frequency deviation and regional control error signals. The terminal combines external disturbances and attack disturbances into a composite disturbance and, based on preset proportional-integral control information, constructs a model containing... A closed-loop control system model is constructed using both frequency control and deception attack models. The terminal constructs a Lyapunov-Krasovsky functional to perform stability analysis on the closed-loop control system model. The Lyapunov-Krasovsky functional includes current state terms, delayed state terms, historical derivative terms, and game-theoretic constraint terms. The terminal derives the functional derivative to determine the constraints that ensure the closed-loop control system model meets preset robust performance indicators. These constraints are stability conditions that satisfy linear matrix inequalities, unifying stability constraints and disturbance suppression performance indicators within the same analytical framework. The terminal uses a bisection method to search for the minimum feasible disturbance suppression level. Based on the constraints, the Grey Wolf optimization algorithm is used to search and determine the target proportional-integral controller gain information. The Grey Wolf optimization algorithm selects the best-performing individual for position update based on individual feasibility judgment and performance indicator calculation results. The terminal performs frequency control on the power system based on the target proportional-integral controller gain information to effectively suppress frequency deviations.

[0089] In the aforementioned frequency control method for integrated energy storage to resist spoofing attacks in power networks, the following steps are taken: First, system operating parameters of the power system including energy storage units are obtained. Based on these parameters, a frequency control model is constructed, incorporating dynamic characteristic information of both generator sets and the energy storage system. Second, a spoofing attack model targeting the frequency control model is constructed based on preset frequency-driven spoofing attack information. Third, a closed-loop control system model including both the frequency control model and the spoofing attack model is constructed based on preset PI control information. Fourth, stability analysis is performed on the closed-loop control system model to determine constraints that ensure it meets preset robust performance indicators. Fifth, the target PI controller gain information is determined based on these constraints. Finally, frequency control of the power system is performed based on the target PI controller gain information. This scheme acquires the system operating parameters of the power system including energy storage units and constructs a frequency control model that includes dynamic characteristic information of generator sets and energy storage systems, which is beneficial for comprehensively characterizing the dynamic characteristics of the power system. By constructing a deception attack model targeting the frequency control model and establishing a closed-loop control system model, it is beneficial for incorporating attack behavior into the control system analysis framework. By performing stability analysis on the closed-loop control system model and determining the constraints that meet the preset robust performance indicators, it is beneficial for ensuring the stability of the system under attack conditions. By determining the target proportional-integral controller gain information based on the constraints and performing frequency control, it is beneficial for the power system to maintain efficient and robust frequency regulation performance even when subjected to deception attacks, thereby improving the frequency control stability of the power system.

[0090] In an exemplary embodiment, a deception attack model targeting a frequency control model is constructed based on preset frequency-driven deception attack information, including: determining attack weight parameters based on the frequency-driven deception attack information; performing attack perturbation processing on the frequency information in the frequency control model based on the attack weight parameters to obtain damaged frequency information; performing attack perturbation processing on the regional control error signal in the frequency control model based on the attack weight parameters to obtain damaged regional control error signal; and constructing a deception attack model based on the damaged frequency information and the damaged regional control error signal.

[0091] The attack weight parameter can be a time-varying attack weight matrix function compatible with the dimension of the identity matrix. The attack weight parameter is used to describe the strength and type of fake data injection attack. When the attack weight parameter takes different values, it can represent scaling attack or perturbation attack.

[0092] Among them, frequency information can be the frequency deviation measurement signal in the frequency control model. Frequency information is a key measurement signal in the power system that is vulnerable to false data injection attacks.

[0093] Among them, attack perturbation processing can be a process of applying attack weight parameters to the real measurement signal and injecting bounded forged data signals into the real measurement signal.

[0094] The damaged frequency information can be the actual measured value of the frequency deviation under a fake data injection attack. The damaged frequency information is composed of the superposition of the real frequency deviation and the injected bounded fake data signal.

[0095] Among them, the area control error signal can be the product of the frequency deviation factor and the frequency deviation. The area control error signal is an important feedback signal in the load frequency control system.

[0096] Among them, the damaged area control error signal can be the actual measured value of the area control error under a false data injection attack. The damaged area control error signal is composed of the superposition of the real area control error signal and the disturbance signal caused by the attack.

[0097] Optionally, the terminal determines attack weight parameters based on frequency-driven deception attack information. The attack weight parameters are time-varying attack weight matrices. Based on the attack weight parameters, the terminal performs attack perturbation processing on the frequency information in the frequency control model, injects bounded forged data signals into the frequency deviation measurement signal, and obtains damaged frequency information. Based on the attack weight parameters, the terminal performs attack perturbation processing on the regional control error signal in the frequency control model, and obtains damaged regional control error signals. Based on the damaged frequency information and the damaged regional control error signals, the terminal constructs a deception attack model that can reflect the attack strategy and impact path.

[0098] The technical solution provided in this embodiment determines the attack weight parameters based on frequency-driven deception attack information, and performs attack perturbation processing on frequency information and regional control error signals based on the attack weight parameters. This is beneficial for comprehensively characterizing the attacker's tampering behavior with key measurement signals, thereby facilitating the construction of a deception attack model that can reflect real attack scenarios.

[0099] In an exemplary embodiment, stability analysis of the closed-loop control system model includes: constructing a game model based on controller information and attacker information; determining a target performance index based on the game model, wherein the target performance index is used to characterize the controller's ability to suppress disturbance signals under the most unfavorable attack conditions; and performing stability analysis on the closed-loop control system model based on the target performance index.

[0100] The controller information can be the control strategy parameter information of the proportional-integral controller. The controller, as the defender, resists the attack by minimizing the performance indicators.

[0101] Among them, attacker information can be adversarial perturbation signals injected by the attacker and descriptions of attack strategies. The attacker aims to undermine system stability by maximizing performance metrics.

[0102] Among them, the game model can be a mathematical model that models the controller and the attacker as a two-person zero-sum differential game relationship. The game model unifies the modeling of external disturbances and intelligent adversarial signals.

[0103] The target performance index can be a minimum-maximum performance index. The target performance index includes a state weighting term, a control input weighting term, and a disturbance suppression term. The target performance index is used to measure the system's performance under the most unfavorable attack conditions.

[0104] Optionally, the terminal constructs a game model based on controller information and attacker information. The game model models the adversarial relationship between the controller and the attacker as a two-player zero-sum differential game problem. Based on the game model, the terminal determines the target performance index, which is a minimum-maximum performance index. The target performance index includes a state weighting matrix, a control input weighting matrix, and a disturbance suppression level parameter. The target performance index is used to characterize the controller's ability to suppress disturbance signals under the most unfavorable attack conditions. Based on the target performance index, the terminal performs stability analysis on the closed-loop control system model, transforming the system control problem into an optimal control problem containing intelligent game disturbance terms.

[0105] The technical solution provided in this embodiment constructs a game model based on controller information and attacker information, and determines the target performance index based on the game model. This is beneficial for considering the adversarial relationship between the controller and the attacker from a game theory perspective, thereby facilitating the robust stability analysis of the system under the most unfavorable attack conditions.

[0106] In an exemplary embodiment, determining the target PI controller gain information according to constraints includes: initializing a candidate gain parameter population within a preset gain parameter range; performing fitness identification processing on each candidate gain parameter in the candidate gain parameter population according to constraints to obtain the fitness value of each candidate gain parameter; performing iterative update processing on the candidate gain parameter population according to the fitness value of each candidate gain parameter to obtain the target gain parameter; and determining the target PI controller gain information according to the target gain parameter.

[0107] The preset gain parameter range can be the upper and lower bounds of the pre-defined proportional gain parameter and integral gain parameter.

[0108] The candidate gain parameter population can be a set of multiple candidate controller gain parameters randomly initialized within a preset gain parameter range.

[0109] Among them, fitness identification processing can be a process of determining the feasibility of candidate gain parameters and calculating performance indicators based on constraints.

[0110] Among them, the fitness value can be used as an evaluation index to measure the degree to which candidate gain parameters meet the constraints and the quality of their performance.

[0111] The iterative update process can be a process of selecting candidate gain parameters with excellent performance based on fitness values, and guiding the other individuals to update their positions based on the positions of the excellent individuals.

[0112] The target gain parameter can be the optimal gain parameter obtained after multiple iterative searches. The target gain parameter makes the closed-loop system satisfy the stability condition and performance constraints of the linear matrix inequality.

[0113] Optionally, within a preset gain parameter range, the terminal initializes a candidate gain parameter population, which includes multiple randomly generated combinations of candidate proportional gain parameters and candidate integral gain parameters. Based on constraints, the terminal performs fitness identification processing on each candidate gain parameter in the candidate gain parameter population, determines whether each candidate gain parameter satisfies the stability condition of the linear matrix inequality, calculates performance indicators, and obtains the fitness value of each candidate gain parameter. Based on the fitness values ​​of each candidate gain parameter, the terminal selects the candidate gain parameter with excellent performance, iteratively updates the candidate gain parameter population, obtains the target gain parameter, and determines the target proportional-integral controller gain information based on the target gain parameter.

[0114] The technical solution provided in this embodiment initializes a population of candidate gain parameters within a preset gain parameter range and performs fitness identification and iterative update processing according to constraints. This facilitates the global search for optimal controller parameters while satisfying stability constraints, thereby helping to obtain target proportional-integral controller gain information with optimal disturbance suppression performance.

[0115] In an exemplary embodiment, a frequency control model containing dynamic characteristic information of the generator set and dynamic characteristic information of the energy storage system is constructed based on system operating parameters. This includes: determining the power balance information of the power system based on the system operating parameters; determining state space information based on the power balance information, wherein the state variables in the state space information include frequency deviation, governor valve position deviation, turbine mechanical power deviation, and energy storage unit output power deviation; and constructing the frequency control model based on the state space information.

[0116] Among them, power balance information can be the dynamic relationship information of load frequency control established based on the power balance equation. Power balance information reflects the balance relationship between the power output of generator sets and energy storage systems and the load demand.

[0117] Among them, state-space information can be system description information that represents the dynamic characteristics of the power system in the form of state-space equations.

[0118] Among them, frequency deviation can be the difference between the actual frequency and the rated frequency. Frequency deviation is the core monitoring variable of the load frequency control system.

[0119] Among them, the governor valve position deviation can be the difference between the actual position and the set position of the governor valve.

[0120] Among them, the turbine mechanical power deviation can be the difference between the actual mechanical power output of the turbine and the set power.

[0121] Among them, the output power deviation of the energy storage unit can be the difference between the actual output power of the energy storage unit and the set power.

[0122] Optionally, the terminal determines the power balance information of the power system based on the system operating parameters. The power balance information is established based on the dynamic characteristics of the main generator, governor, turbine, and energy storage system. Based on the power balance information, the terminal determines the state space information. The state variables in the state space information include frequency deviation, governor valve position deviation, turbine mechanical power deviation, and energy storage unit output power deviation. Based on the state space information, the terminal constructs a frequency control model, which expresses the dynamic characteristics of the power system in the form of state space equations.

[0123] The technical solution provided in this embodiment determines power balance information and state space information based on system operating parameters, which is beneficial for comprehensively characterizing the dynamic response characteristics of generator sets and energy storage systems, thereby facilitating the construction of a frequency control model that can accurately reflect the dynamic behavior of the power system.

[0124] In an exemplary embodiment, the closed-loop control system model includes composite disturbance information; the method further includes: acquiring external disturbance information, determining attack disturbance information according to a deception attack model, and fusing the external disturbance information and the attack disturbance information to obtain composite disturbance information.

[0125] Among them, composite disturbance information can be a unified disturbance description information formed by combining external disturbance information and attack disturbance information.

[0126] Among them, external disturbance information can be disturbance signal information caused by factors such as sudden external interference and communication network noise.

[0127] Among them, attack perturbation information can be bounded perturbation signal information caused by intelligent adversarial signals, and attack perturbation information is used to describe the attacker's strategy input.

[0128] Among them, fusion processing can be a process of combining external disturbance information and attack disturbance information according to preset rules.

[0129] Optionally, the terminal acquires external disturbance information, which represents disturbances caused by sudden external interference. Based on the deception attack model, the terminal determines attack disturbance information, which represents bounded disturbances caused by intelligent countermeasure signals. The terminal performs fusion processing on the external disturbance information and the attack disturbance information, combining them into composite disturbance information. The closed-loop control system model includes composite disturbance information to facilitate consideration of the impact of multiple disturbance factors within the same analysis framework.

[0130] The technical solution provided in this embodiment obtains external disturbance information and determines attack disturbance information according to the deception attack model. It then fuses the external disturbance information and the attack disturbance information to obtain composite disturbance information. This is beneficial for considering multiple disturbance factors in a unified manner in the closed-loop control system model, thereby improving the integrity of the closed-loop control system model and the comprehensiveness of the analysis.

[0131] The following example illustrates the frequency control method for integrated energy storage and anti-spoofing attacks for power networks provided in this application. This example demonstrates the application of this method to a terminal.

[0132] With the deep penetration of new energy sources into the power system, system inertia has decreased significantly, making traditional frequency regulation mechanisms dominated by large synchronous machines insufficient to meet the dynamic stability requirements of complex power grids. While the introduction of energy storage systems enhances the flexibility of system frequency regulation, it also makes the control loop structure more complex, significantly increasing time delay effects and multi-source interference. Traditional fixed-parameter PI controllers are prone to problems such as response hysteresis, increased overshoot, and prolonged recovery time under dynamic disturbances and nonlinear effects, making it difficult to achieve high-precision frequency stability. Simultaneously, the networked and information-based operation of the power system exposes it to the potential threats of "heterogeneous and complex power networks." Frequency measurement, control signals, and energy dispatch are highly susceptible to attacks by attackers attempting "deception probing" and "fictitious data injection (FDI)." Attackers can forge or delay the transmission of control data to achieve malicious traffic injection and data deception in cross-domain communication channels, thereby disrupting the stable operation of the system.

[0133] To address this issue, existing research largely relies on attack detection or intrusion trapping systems for passive defense. However, when faced with highly covert and dynamically evolving attack scenarios, these methods often suffer from problems such as "insufficient trapping and identification accuracy" and "lagging defense response," failing to guarantee the robustness and security of frequency control in real time during an attack. Particularly in "multi-source heterogeneous network" environments, the complex attack paths and intertwined event chains make "spatiotemporal tracing and attack reproduction" of the control system difficult to achieve, severely restricting the security, controllability, and post-event analysis capabilities of the frequency modulation system.

[0134] To address the aforementioned issues, this embodiment studies control security mechanisms for power system network attack and defense, proposing a PI frequency control method for integrated energy storage power systems to resist deception and false data attacks based on game optimization. This method models the controller and attacker as a dynamic game relationship, introducing a three-layer structure of "attack-defense-entrapment" into the system. Attack behavior, entrapment characteristics, and control parameter optimization are incorporated into the same decision framework, achieving proactive defense and adaptive parameter adjustment. By constructing an FDI deception model with time-varying attack weights and combining it with a genetic algorithm for global optimization search of the PI gain parameters, the controller can maintain efficient and robust frequency regulation performance even under conditions of multi-domain communication disturbances and intertwined deception data.

[0135] In summary, this embodiment aims to solve the frequency security control problem of integrated energy storage power systems under network attack conditions, and to construct a PI control method that combines active game optimization, adaptive adjustment and robust defense characteristics, so as to provide sustainable technical support for the safe and stable operation of complex networked power systems.

[0136] This embodiment addresses the technical challenge of integrated energy storage power systems being vulnerable to deceptive probing and fictitious data injection (FDI) attacks in heterogeneous network environments. It proposes a PI frequency control method based on game theory optimization. This method combines the dynamic response characteristics of energy storage with network attack-defense game mechanisms. By introducing a time-varying deception weight model, an attack-defense two-layer decision structure, and a genetic algorithm optimization strategy, it achieves proactive and robust control of the system under cross-domain communication disturbances and deceptive attacks. The overall process includes the following steps:

[0137] S1: Establish a frequency control model for an integrated energy storage power system.

[0138] S2: Construct a fake data injection and deceptive probing attack model.

[0139] S3: Establish an optimization framework for frequency control in attack and defense game theory.

[0140] S4: Design a PI controller parameter optimization method based on the gray wolf algorithm.

[0141] This embodiment first focuses on a power system containing energy storage units, considering time-varying transmission delays caused by communication networks and pulse disturbances caused by sudden external interference, and establishes a state-space mathematical model of the system. This model treats the main generator, governor, turbine, and energy storage system as a unified dynamic object, and establishes the dynamic relationship of load frequency control (LFC) through power balance equations. The system model is represented as follows:

[0142] (1)

[0143] in , , , , and These are frequency deviation, frequency preset value deviation, turbine valve position deviation, generator mechanical power deviation, load deviation, and power output deviation. , , , , , and These represent the equivalent inertial constant, damping coefficient, turbine time constant, governor time constant, droop coefficient, energy storage unit time constant, and unit adjustment factor, respectively. and Let be the participation ratios of the speed governor and the energy storage system, respectively, and satisfy the condition that their sum is 1. Let , , , .in It is a regional control error signal. It is the frequency deviation factor. Based on the system model (1) and considering the influence of pulse interference, the state-space equation model of the system can be established as follows:

[0144] (2)

[0145] In practical engineering applications, attackers must consider both the stealth and effectiveness of their attack signals. This requires attack signals to be dynamically adjusted to evade detection by signal detectors and to launch attacks efficiently at the optimal time to maximize impact.

[0146] As a further improvement to this embodiment, step 2 is as follows:

[0147] To improve the security performance of the LFC system, this embodiment proposes an FDI attack signal model from the attacker's perspective, incorporating a time-varying attack weight matrix. This method helps in designing more targeted proactive defense strategies. FDI attacks in the LFC system can be modeled as follows:

[0148] (3)

[0149] in, This indicates the signal actually received by the controller. Represents the actual sampled signal. For the identity matrix, its dimensions are... Reflects the signal The characteristics; To and Dimensionally compatible FDI attack weight matrix function. When When different values ​​are taken, FDI can represent different types of attacks.

[0150] when When the matrix is ​​scalar, the corresponding scaling attack is:

[0151] (4)

[0152] in, , Representation matrix Elements on the diagonal, and .when When it is a general matrix, it corresponds to a perturbation attack.

[0153] In power systems, the measurement signals most vulnerable to FDI attacks primarily include frequency signals. Therefore, under an FDI attack, the compromised measurement values ​​of an LFC system can be expressed as:

[0154] (5)

[0155] in, This represents the actual measured value of the frequency deviation under an FDI attack. This refers to a bounded, forged data signal injected into a frequency signal.

[0156] Under an FDI attack, the Area Control Error (ACE) signal is defined as follows:

[0157] (6)

[0158] in, To characterize the adversarial perturbation caused by FDI attacks, a bounded perturbation signal is defined. This represents the attacker's policy input, assuming... The set of disturbances is: External disturbances With attack disturbance The combination results in a complex disturbance: ,in, Indicates external noise. This represents a smart adversarial signal. Based on the aforementioned FDI attack and complex perturbation modeling, a game theory performance optimization model is further constructed.

[0159] As a further improvement to this embodiment, step 3 is as follows:

[0160] In order to simultaneously consider external disturbances in controller design Signals of countering intelligentization To mitigate the impact of the attack, this embodiment models both issues from a game theory perspective as a two-player zero-sum differential game. The controller, acting as the defender, minimizes the performance metric to resist the attacker's damage to the system performance under the most unfavorable conditions. Specifically, the minimum-maximum performance metrics are set as follows:

[0161] (7)

[0162] in, and These are the weighted matrices for the state and control inputs, respectively. The disturbance suppression level is; the control input is... Within this framework, the fuzzy controller design problem can be transformed into the following two-person zero-sum optimal control problem:

[0163] (8)

[0164] By solving the above optimization problem under the premise of satisfying the Lyapunov stability condition, a fuzzy controller gain that can still maintain the closed-loop stability of the system and has robust performance under the most unfavorable attack scenario can be obtained.

[0165] Preset system controller gain ,and For a given , as well as If there exists a matrix of appropriate dimension such that the following LMI holds, then the power system is globally asymptotically stable:

[0166] (9)

[0167] in,

[0168]

[0169]

[0170]

[0171]

[0172] , ,

[0173] ,

[0174] .

[0175] To verify the asymptotic stability of the system, the following Lyapunov-Krasovskii functional is constructed:

[0176] (10)

[0177] Each sub-function in this formula Corresponding to the current state term, the delayed state term, the historical derivative term, and the game constraint term, respectively, the overall derivative is expressed as:

[0178] (11)

[0179] After the above derivation, in order to verify the stability of the evaluation system, for the matrix... integer and And satisfy If any continuous-time variable x is well defined, then:

[0180] (12)

[0181] in:

[0182] ,

[0183] External interference Under the given conditions, the derivative of the system's Lyapunov function can ultimately satisfy:

[0184] As a further improvement to this embodiment, it also includes:

[0185] (13)

[0186] in,

[0187] like Then it can be deduced that This ensures the asymptotic stability of the entire energy storage system.

[0188] In an adversarial operating environment, This is considered an adversarial disturbance signal injected by an intelligent adversary. The control objective of this embodiment is to minimize the impact of this disturbance on system performance under the worst-case scenario while ensuring the stability of the closed-loop system.

[0189] Therefore, a minimum-maximum performance criterion is set, requiring that under all acceptable disturbances... Under the given conditions, the system's infinite time-domain performance index is bounded, that is:

[0190] (14)

[0191] in, Let be the initial value of the candidate Lyapunov function, which is zero.

[0192] To ensure that the above inequality holds, the following differential inequality must be satisfied on the system trajectory:

[0193] (15)

[0194] Based on this, the following inequalities can be derived:

[0195] (16)

[0196] This condition unifies stability constraints and disturbance suppression performance indicators within the same analytical framework, providing a feasible basis for the design of robust fuzzy controllers based on game theory performance guarantees.

[0197] Within the proposed game theory framework, the level of perturbation inhibition This plays a crucial role in the robustness of the closed-loop system. To achieve the optimal balance between disturbance suppression capability and controller feasibility, this embodiment employs a bisection method to search for the minimum feasible solution. This allows the system to simultaneously satisfy stability and performance constraints under a given controller gain.

[0198] As a further improvement to this embodiment, step 4 is as follows:

[0199] To obtain the minimum feasible disturbance suppression level, the Grey Wolf optimization algorithm is used to search for the optimal fuzzy controller gain vector under this constraint.

[0200] Specifically, the gray wolf population is first randomly initialized within a given upper and lower bound interval. Then, in each iteration, the three best-performing gray wolves are selected based on individual feasibility assessment and performance index calculation results, and their positions guide the updating of the remaining individuals.

[0201] The updated formula is shown below:

[0202] (17)

[0203] in, The coefficient is a linearly decreasing factor. and The variable is a uniform random number within the interval [0, 1]; clip() represents the boundary clipping operation, used to ensure that the controller gain is within a preset range. Through multiple iterative searches, the optimal gain vector is finally obtained, so that the LMI stability condition and performance constraints of the closed-loop system are satisfied while achieving the optimal disturbance suppression performance.

[0204] This embodiment addresses the state deviation and control interference problems caused by frequency-driven FDI attacks in power systems containing energy storage units. A unified, generalized state-space modeling framework is constructed, comprehensively considering communication network delay, disturbance uncertainty, and the time-varying characteristics of attack signals. By introducing frequency deviation and ACE signal tampering modeling methods, a system dynamic model reflecting attack strategies and impact paths is established. To improve the system's ability to identify different types of attacks and its anti-interference performance, a disturbance observer and a composite disturbance modeling mechanism are designed, further transforming the system control problem into a two-person zero-sum optimal control problem with intelligent game interference terms.

[0205] Based on the aforementioned modeling foundation, to solve for the optimal controller parameters of the system under the most unfavorable attack scenario, this embodiment employs a robust stability analysis method using the Lyapunov-Krasovskii function as a tool to derive the LMI condition that satisfies robust stability and performance index constraints. Subsequently, under this constraint, the Grey Wolf optimization algorithm is constructed, combining a three-optimal leading mechanism and a position update strategy to search for the optimal fuzzy controller gain, thereby enhancing the robustness and anti-disturbance capability of the controller while balancing attack resistance and dynamic performance.

[0206] In summary, this embodiment achieves robust controller design for power systems with energy storage under frequency-driven attacks through a systematic approach integrating attack modeling, stability assessment, and performance search. This approach not only effectively suppresses the risk of frequency drift caused by intelligent interference but also significantly improves the system's stability and safety margin under the most unfavorable conditions, demonstrating significant technical advantages such as clear structure, strong anti-interference capability, and good adaptability.

[0207] To verify the feasibility and superiority of the method in this embodiment, a typical single-area power system with energy storage units was used as the simulation object for testing. The system was equipped with conventional generators and supercapacitor energy storage devices, employing LFC control to achieve frequency stability. The main parameters of the simulation system are shown in Table 1, including the inertial parameters of the conventional units and energy storage devices, the governor time constant, the proportional coefficient, and the frequency regulation coefficient, all of which were selected as typical values. During the simulation, a frequency-driven FDI attack signal was injected into the area to dynamically disturb the frequency measurements.

[0208] The comparison schemes include the following two: 1) traditional PI controller; 2) the controller design scheme proposed in this embodiment.

[0209] This embodiment compares and analyzes the frequency response and performance indicators of the traditional PI controller and the proposed Robust PI (RPI) controller under typical disturbance conditions. In the initial disturbance phase, the RPI controller significantly reduces the maximum overshoot of the frequency deviation and exhibits faster response speed and a smoother dynamic adjustment process. Even under subsequent disturbances or attacks, the RPI controller maintains good robustness, demonstrating stronger anti-disturbance and frequency recovery capabilities. In terms of quantitative performance indicators, the traditional PI controller has an integral time absolute error (ITAE) of 0.191199 and an integral squared error (ISE) of 0.000016, while the RPI controller reduces its ITAE to 0.151112 and ISE to 0.000030. Although the RPI controller has a slightly higher integral squared error, it reduces the ITAE, a measure of dynamic response quality, by approximately 21%, significantly improving the frequency control performance of the system throughout its operation. In summary, the RPI control strategy proposed in this embodiment outperforms traditional methods in terms of disturbance suppression, dynamic performance, and system robustness, and has good engineering feasibility and value.

[0210] Figure 2 This diagram presents a load frequency control topology for a power system containing energy storage units. It includes modules such as a sliding mode controller, a zero-order hold, a governor, a turbine, a non-reheat gas turbine, an energy storage system, and a generator. The transfer function of the governor is 1 divided by (governor time constant multiplied by s plus 1), the turbine's transfer function is 1 divided by (turbine time constant multiplied by s plus 1), the energy storage system's transfer function is the energy storage unit participation factor divided by (energy storage unit time constant multiplied by s plus 1), and the generator's transfer function is 1 divided by (equivalent inertia constant multiplied by s plus damping coefficient). 's' represents the complex frequency variable in the Laplace transform, the droop coefficient is a frequency regulation characteristic parameter, 'R' represents the droop coefficient used for frequency regulation characteristics, and the frequency deviation is the system output. It also includes a frequency deviation factor, a governor participation proportionality coefficient, an energy storage system participation proportionality coefficient, and an energy storage unit output power deviation.

[0211] Figure 3 The frequency deviation graph shows the frequency deviation under different control strategies. The horizontal axis represents time in seconds, ranging from 0 to 120 seconds, and the vertical axis represents the frequency deviation, ranging from -0.5 × 10⁻⁶ seconds. -3 Up to 3×10 -3 The curves in the figure represent the frequency response curves of a conventional proportional-integral (PI) controller and a robust proportional-integral (RPI) controller, respectively.

[0212] Table 1 (System Parameters)

[0213]

[0214] Table 1 shows the system parameters: the time constant of the turbine is 0.10, the time constant of the governor is 0.30, the time constant of the energy storage unit is 0.10, the participation ratios of the governor and the energy storage system are 0.50 and 0.50, respectively, the frequency deviation factor is 21, the equivalent inertia constant is 10.00, the damping coefficient is 1.00, the droop coefficient is 0.05, the unit adjustment factor is 1 / 21, and the energy storage unit participation factor is 1.00.

[0215] A power system with energy storage units based on a traditional proportional-integral (PI) strategy was compared with a power system with energy storage units based on a robust PI strategy. The dynamic performance indicators included the absolute integral time error and the integral square error. The absolute integral time error of the traditional PI strategy was 0.191199 and the integral square error was 0.000016, while the absolute integral time error of the robust PI strategy was 0.151112 and the integral square error was 0.000030.

[0216] Compared to existing robust control methods for load frequency control, this embodiment constructs a complete defense closed-loop system from attack modeling and controller design to performance optimization. While ensuring the frequency stability of the power system, it significantly improves the system's anti-attack capability and adaptive control level under harsh network conditions.

[0217] Specifically, this embodiment innovatively introduces a frequency-driven deception attack model (FDI), which can characterize the attacker's ability to dynamically adjust the attack strategy based on the system's observed signals, thereby effectively reflecting the complex attack behaviors in reality that are highly concealed and have varied strategies. At the same time, by establishing a state feedback framework that includes control signals, measurement information, and composite disturbance terms, and combining generalized linear system modeling and Lyapunov-Krasovskii function design, it ensures that the system still meets the stability requirements when facing unknown disturbances and intentionally tampered signals.

[0218] In particular, considering the increasingly heterogeneous power system network structure and diversified service paths, the method in this embodiment possesses good cross-domain adaptability and scalability. In multi-regional wide-area power control, complex deception behaviors such as redundant channel data spoofing, gateway-level attack penetration, and forged control responses often exist, making it difficult for traditional methods to achieve continuous and accurate identification and response. This embodiment, by combining controller optimization with countermeasure strategy design, can be effectively embedded into a heterogeneous and complex power network deception detection system, providing control-side foundational support for subsequent advanced detection mechanisms such as cross-domain malicious traffic detection and dynamic game-theoretic strategy identification.

[0219] From a performance perspective, simulation results show that under typical disturbance and spoofing attack scenarios, the peak frequency deviation of the traditional PI controller reaches 2.6 × 10⁻⁶. -3 The frequency deviation is 0.191199 Hz (Hertz), while the RPI controller proposed in this embodiment suppresses the frequency deviation to 1.5 × 10⁻⁶. -3 The ITAe value dropped to 0.151112 Hz, a decrease of more than 21%. This result proves that this embodiment not only improves the controller's immunity to disturbances, but also significantly enhances stable control capabilities in scenarios involving deception attack detection and policy switching.

[0220] In summary, this embodiment possesses superior dynamic control capabilities, frequency stability, and attack resistance capabilities. Especially in heterogeneous and complex power network environments, it can effectively support redundant service detection and strategy defense, demonstrating high engineering practical value and network security.

[0221] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0222] Based on the same inventive concept, this application also provides a frequency control device for integrated energy storage and anti-spoofing attacks on power networks, which implements the frequency control method for integrated energy storage and anti-spoofing attacks on power networks described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the frequency control device for integrated energy storage and anti-spoofing attacks on power networks provided below can be found in the limitations of the frequency control method for integrated energy storage and anti-spoofing attacks on power networks described above, and will not be repeated here.

[0223] In one exemplary embodiment, such as Figure 4 As shown, a frequency control device 400 integrating energy storage and anti-spoofing attacks for power networks is provided. This device may include:

[0224] The parameter acquisition module 401 is used to acquire the system operating parameters of the power system containing the energy storage unit, and to construct a frequency control model containing the dynamic characteristic information of the generator set and the dynamic characteristic information of the energy storage system based on the system operating parameters.

[0225] The model building module 402 is used to build a deception attack model against the frequency control model based on preset frequency-driven deception attack information, and to build a closed-loop control system model including the frequency control model and the deception attack model based on preset PI control information.

[0226] The model analysis module 403 is used to perform stability analysis on the closed-loop control system model and determine the constraints that make the closed-loop control system model meet the preset robust performance index.

[0227] Information determination module 404 is used to determine the target PI controller gain information based on the constraints.

[0228] The system control module 405 is used to perform frequency control on the power system based on the target PI controller gain information.

[0229] In an exemplary embodiment, the model building module 402 is further configured to: determine attack weight parameters based on frequency-driven deception attack information; perform attack perturbation processing on the frequency information in the frequency control model based on the attack weight parameters to obtain damaged frequency information; perform attack perturbation processing on the regional control error signal in the frequency control model based on the attack weight parameters to obtain damaged regional control error signal; and construct a deception attack model based on the damaged frequency information and the damaged regional control error signal.

[0230] In an exemplary embodiment, the model analysis module 403 is further configured to construct a game model based on controller information and attacker information; determine a target performance index based on the game model; the target performance index is used to characterize the controller's ability to suppress disturbance signals under the most unfavorable attack conditions; and perform stability analysis processing on the closed-loop control system model based on the target performance index.

[0231] In an exemplary embodiment, the information determination module 404 is further configured to initialize a candidate gain parameter population within a preset gain parameter range; perform fitness identification processing on each candidate gain parameter in the candidate gain parameter population according to the constraint conditions to obtain the fitness value of each candidate gain parameter; perform iterative update processing on the candidate gain parameter population according to the fitness value of each candidate gain parameter to obtain the target gain parameter; and determine the target PI controller gain information according to the target gain parameter.

[0232] In an exemplary embodiment, the parameter acquisition module 401 is further configured to determine the power balance information of the power system based on the system operating parameters; determine the state space information based on the power balance information; the state variables in the state space information include frequency deviation, governor valve position deviation, turbine mechanical power deviation and energy storage unit output power deviation; and construct a frequency control model based on the state space information.

[0233] In an exemplary embodiment, the closed-loop control system model includes composite disturbance information; the device 400 further includes: an information fusion module for acquiring external disturbance information, determining attack disturbance information according to a deception attack model, and fusing the external disturbance information and the attack disturbance information to obtain composite disturbance information.

[0234] The modules in the aforementioned frequency control device for integrated energy storage and anti-spoofing attacks for power networks can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0235] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 5 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a frequency control method for integrated energy storage and anti-spoofing attacks for power grids. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0236] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0237] In one exemplary embodiment, a computer device is also 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 in the above-described method embodiments.

[0238] In one exemplary embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above-described method embodiments.

[0239] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0240] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0241] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0242] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A frequency control method for integrated energy storage and anti-spoofing attacks in power networks, characterized in that, The method includes: Obtain the system operating parameters of the power system containing energy storage units, and construct a frequency control model containing dynamic characteristic information of generator sets and dynamic characteristic information of energy storage systems based on the system operating parameters; Based on preset frequency-driven deception attack information, a deception attack model targeting the frequency control model is constructed, and based on preset PI control information, a closed-loop control system model including the frequency control model and the deception attack model is constructed. The closed-loop control system model is subjected to stability analysis to determine the constraints that enable the closed-loop control system model to meet the preset robust performance index. Based on the constraints, determine the target PI controller gain information; Frequency control of the power system is performed based on the target PI controller gain information.

2. The method according to claim 1, characterized in that, The step of constructing a deception attack model targeting the frequency control model based on preset frequency-driven deception attack information includes: Based on the frequency-driven spoofing attack information, determine the attack weight parameters; Based on the attack weight parameters, the frequency information in the frequency control model is subjected to attack perturbation processing to obtain damaged frequency information; Based on the attack weight parameters, the regional control error signal in the frequency control model is subjected to attack perturbation processing to obtain the damaged regional control error signal; The deception attack model is constructed based on the damaged frequency information and the damaged area control error signal.

3. The method according to claim 1, characterized in that, The stability analysis of the closed-loop control system model includes: Construct a game theory model based on controller and attacker information; Based on the game theory model, a target performance index is determined; the target performance index is used to characterize the controller's ability to suppress disturbance signals under the most unfavorable attack conditions. Based on the target performance indicators, the closed-loop control system model is subjected to stability analysis.

4. The method according to claim 1, characterized in that, Determining the target PI controller gain information based on the constraints includes: Within the preset gain parameter range, initialize the candidate gain parameter population; Based on the constraints, fitness identification processing is performed on each candidate gain parameter in the candidate gain parameter population to obtain the fitness value of each candidate gain parameter. Based on the fitness values ​​of each candidate gain parameter, the candidate gain parameter population is iteratively updated to obtain the target gain parameter; The target PI controller gain information is determined based on the target gain parameter.

5. The method according to claim 1, characterized in that, The step of constructing a frequency control model that includes dynamic characteristic information of the generator set and dynamic characteristic information of the energy storage system based on the system operating parameters includes: Based on the system operating parameters, determine the power balance information of the power system; Based on the power balance information, state space information is determined; the state variables in the state space information include frequency deviation, governor valve position deviation, turbine mechanical power deviation, and energy storage unit output power deviation. The frequency control model is constructed based on the state space information.

6. The method according to any one of claims 1 to 5, characterized in that, The closed-loop control system model includes composite disturbance information; The method further includes: Obtain external disturbance information and determine the attack disturbance information based on the deception attack model; The external disturbance information and the attack disturbance information are fused to obtain the composite disturbance information.

7. A frequency control device integrating energy storage and anti-spoofing attacks for power networks, characterized in that, The device includes: The parameter acquisition module is used to acquire the system operating parameters of the power system containing the energy storage unit, and to construct a frequency control model containing the dynamic characteristic information of the generator set and the dynamic characteristic information of the energy storage system based on the system operating parameters. The model building module is used to build a deception attack model against the frequency control model based on preset frequency-driven deception attack information, and to build a closed-loop control system model containing the frequency control model and the deception attack model based on preset PI control information. The model analysis module is used to perform stability analysis on the closed-loop control system model and determine the constraints that make the closed-loop control system model meet the preset robust performance index. The information determination module is used to determine the target PI controller gain information based on the constraints. The system control module is used to perform frequency control on the power system based on the target PI controller gain information.

8. 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 method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images