Power grid system control method and device considering deterministic network communication attack

Abstract

The embodiment of the invention provides a power grid system control method and device considering deterministic network communication attack. The method comprises the following steps: establishing a control model of a direct current micro-grid system; establishing an RBF neural network DC micro-grid prediction model; calculating a deviation value between a monitoring value of the direct-current micro-grid system monitored by the sensor and a prediction value of the RBF neural network direct-current micro-grid prediction model; judging whether a preset event triggering mechanism is triggered or not based on the deviation value and a preset threshold value; if yes, taking a predicted value of the RBF neural network direct current micro-grid prediction model as an input parameter of a super-spiral sliding mode controller; if not, taking the monitored value as an input parameter of the super-spiral sliding mode controller; and according to the input parameters, setting a super-spiral sliding mode controller used for converging the output value of the DC micro-grid system. According to the embodiment of the invention, the problem that the system is unstable when the DC micro-grid system is attacked by network communication can be solved.

Description

Technical Field

[0001] The embodiments of this specification relate to the field of network communications, and in particular, to a power grid system control method and device taking into account deterministic network communication attacks. Background Art

[0002] As an emerging network communication architecture, deterministic networks rely on their core technologies of resource reservation, service guarantee and explicit routing to set high standards for data transmission performance. Under this framework, the information exchange between the controllers of the distributed controlled DC microgrid system is highly dependent on the communication network, so it is more vulnerable to network communication attacks, which may cause the microgrid system to deviate from the rated operating point or even cause power outages. In order to improve the stability of the system, a series of measures need to be taken to address these challenges and ensure that the power grid system operates safely and stably under network communication attacks.

[0003] In the prior art, the Chinese invention patent with application number 202110209790.4 uses a univariate linear regression model to linearly fit the voltage differential and current change, and detects the running data. However, in the actual power grid system, there are generally nonlinear factors such as equipment aging, ambient temperature changes, and load mutations, which will cause the linear model to be unable to accurately reflect the system state. At the same time, this method does not consider the problem of system steady-state deviation caused by network communication attacks. The Chinese invention patent with application number 202210272522.1 constructs a controller based on a distributed sliding mode observer, uses a distributed sliding mode observer to estimate the attack signal, and uses the compensation output by the controller to supplement the distributed two-layer control, thereby realizing bus voltage recovery and current output distribution. However, the performance of the distributed sliding mode observer is highly dependent on its parameter setting. Improper parameter setting will weaken the detection ability of the observer, resulting in false alarms or omissions of attack signals, and will also aggravate the jitter phenomenon of the controller under FDI attacks. At the same time, this method does not isolate the detected network communication attacks, which poses a potential threat to the overall stability and reliability of the system.

[0004] Therefore, there is an urgent need for a power grid system control method that takes deterministic network communication attacks into consideration, which can easily and cost-effectively solve the problem of system instability when a distributed controlled DC microgrid system is attacked by network communication. Summary of the invention

[0005] The purpose of the embodiments of this specification is to provide a power grid system control method and device that takes into account deterministic network communication attacks, so as to solve the problem of system instability when a distributed controlled DC microgrid system is attacked by network communication with low difficulty and low cost.

[0006] To achieve the above objectives, on the one hand, an embodiment of this specification provides a power grid system control method considering deterministic network communication attacks, including:

[0007] Establish the control model of DC microgrid system;

[0008] Based on the control model of the DC microgrid system, a RBF neural network DC microgrid prediction model is established;

[0009] Calculate the deviation between the monitoring value of the DC microgrid system obtained by the sensor monitoring and the predicted value of the RBF neural network DC microgrid prediction model;

[0010] Based on the deviation value and the preset threshold, determining whether a preset event trigger mechanism is triggered, wherein the preset event trigger mechanism indicates that the DC microgrid system is subject to a network communication attack;

[0011] If yes, the predicted value of the RBF neural network DC microgrid prediction model is used as an input parameter of the super spiral sliding mode controller;

[0012] If not, the monitoring value is used as an input parameter of the super-helical sliding mode controller;

[0013] According to the input parameters, a super spiral sliding mode controller is set to converge the output value of the DC microgrid system.

[0014] Preferably, the establishing of the control model of the DC microgrid system further comprises:

[0015] Set up multiple distributed generator sets;

[0016] Any distributed generator set is connected to a busbar through a boost converter, and each busbar is connected through resistance and inductance to form a DC microgrid system;

[0017] A mathematical model of the boost converter in one switching cycle is established as the control model of the DC microgrid system.

[0018] Preferably, the mathematical model of the boost converter in one switching cycle includes:

[0019]

[0020]

[0021]

[0022] Among them, L i,f is the inductance of the i-th boost converter, C i,f is the capacitance of the i-th boost converter, For the i-th boost converter through the inductor L f The current, i i,o 、v i,o are the output current and output voltage of the i-th boost converter in the monitoring values ​​monitored by the sensor, d i is the duty cycle of the i-th boost converter switch, r i,d is the equivalent resistance of the i-th boost converter, V i,n is the output voltage of the distributed generator set corresponding to the i-th boost converter, subscript i=1, 2,…, N, N is the number of boost converters, and t is a discrete time point.

[0023] Preferably, the control model of the DC microgrid system is based on the RBF neural network DC microgrid prediction model, which further includes:

[0024] Establish the RBF neural network DC microgrid voltage initial model and the RBF neural network DC microgrid current initial model respectively;

[0025] Based on the actual input of the mathematical model of the boost converter in one switching cycle under different load changes, the input data are respectively used as the RBF neural network DC microgrid voltage initial model and the RBF neural network DC microgrid current initial model;

[0026] Inputting the input data into the RBF neural network DC microgrid voltage initial model and the RBF neural network DC microgrid current initial model respectively to obtain output data of the two initial models;

[0027] Based on the actual output of the mathematical model of the boost converter in one switching cycle under different load changes and the output data of the two initial models, the RBF neural network DC microgrid voltage initial model and the RBF neural network DC microgrid current initial model are optimized respectively to obtain the RBF neural network DC microgrid voltage prediction model and the RBF neural network DC microgrid current prediction model.

[0028] Preferably, the input data is:

[0029]

[0030] in, is the input data of the RBF neural network DC microgrid voltage initial model, or the input data of the RBF neural network DC microgrid current initial model, For the i-th boost converter through the inductor L f The current, d iis the duty cycle of the switch of the i-th boost converter, subscript i=1,2,…,N, N is the number of boost converters;

[0031] The output data is:

[0032] y i,k =w i,k,1 h i,k,1 +w i,k,2 h i,k,2 +…+w i,k,j h i,k,j +…+w i,k,l h i,k,l ;

[0033]

[0034] Among them, when k = 1, y i,k is the output data of the initial model of DC microgrid voltage using RBF neural network, w i,k,j is the weight value of the jth hidden layer neuron in the RBF neural network DC microgrid voltage initial model, h i,k,j is the output of the jth hidden layer neuron in the RBF neural network DC microgrid voltage initial model, b i,k,j is the width of the Gaussian basis function of the jth hidden layer neuron in the RBF neural network DC microgrid voltage initial model, and b i,k,j >0,c i,k,j =[c i,k,j,1 c i,k,j,2 ] T , c i,k,j is the center point vector value of the jth hidden layer neuron, c i,k,j,m is the mth component of the center point vector value of the jth hidden layer neuron, m = 1, 2, j = 1, 2, ..., l, l represents the number of hidden layer neurons;

[0035] When k = 2, y i,k is the output data of the initial current model of the RBF neural network DC microgrid, w i,k,j is the weight value of the jth hidden layer neuron in the RBF neural network DC microgrid current initial model, h i,k,j is the output of the jth hidden layer neuron in the RBF neural network DC microgrid current initial model, b i,k,j is the width of the Gaussian basis function of the jth hidden layer neuron in the RBF neural network DC microgrid current initial model, and b i,k,j >0,c i,k,j =[c i,k,j,1 c i,k,j,2 ] T , c i,k,jis the center point vector value of the jth hidden layer neuron, c i,k,j,m is the mth component of the center point vector value of the jth hidden layer neuron, m=1, 2, j=1, 2, …, l, l represents the number of hidden layer neurons.

[0036] Preferably, the RBF neural network DC microgrid voltage initial model and the RBF neural network DC microgrid current initial model are optimized based on the actual output of the mathematical model of the boost converter in one switching cycle under different load changes and the output data of the two initial models, respectively, to obtain the RBF neural network DC microgrid voltage prediction model and the RBF neural network DC microgrid current prediction model further include:

[0037] Based on the actual output of the mathematical model of the boost converter in one switching cycle under different load changes and the output data of the two initial models, calculating the error value between the actual output and the output data;

[0038] The Adam gradient descent method is used to optimize the model parameters of the two initial models respectively, and the RBF neural network DC microgrid voltage prediction model and the RBF neural network DC microgrid current prediction model corresponding to the minimized error value are obtained.

[0039] Preferably, the calculating the error value between the actual output and the output data based on the actual output of the mathematical model of the boost converter in one switching cycle under different load changes and the output data of the two initial models further comprises:

[0040] The error value between the actual output and the output data is calculated by the following formula:

[0041]

[0042] Among them, when k = 1, y i,k is the output data of the initial model of RBF neural network DC microgrid voltage, is the actual output voltage in the actual output, E i,k is the error value between the actual output voltage and the output data of the RBF neural network DC microgrid voltage initial model; when k = 2, y i,k is the output data of the initial model of the DC microgrid current using the RBF neural network, is the actual output current in the actual output, E i,k is the error value between the actual output current and the output data of the RBF neural network DC microgrid current initial model.

[0043] Preferably, the preset event trigger mechanism includes: a preset event trigger mechanism for current network communication attacks, and a preset event trigger mechanism for voltage network communication attacks;

[0044] The preset event trigger mechanism for current network communication attacks is characterized by the following formula:

[0045]

[0046] Among them, Θ i,I is the nonlinear time-varying trigger function for current network communication attacks, κ i,1 , κ i,2 , κ i,3 , κ i,4 , are constants, e i,I is the deviation value corresponding to the current when the current network communication is attacked, i i,o is the current output by the i-th boost converter in the monitoring value monitored by the sensor, is the predicted current value of the RBF neural network DC microgrid prediction model, subscript i = 1, 2, ..., N, N is the number of boost converters, is the preset threshold for current network communication attacks, is the kth triggering moment of the ith boost converter for the current network communication attack, is the k+1th triggering moment of the ith boost converter for the current network communication attack, inf{·} is the infimum of the set, and t is a discrete time point;

[0047] The preset event trigger mechanism for voltage network communication attacks is characterized by the following formula:

[0048]

[0049] Among them, Θ i,V is the nonlinear time-varying trigger function for voltage network communication attacks, κ i,5 , κ i,6 , κ i,7 , κ i,8 , is a constant, e i,V is the voltage deviation value corresponding to the voltage network communication attack, v i,o is the voltage output by the i-th boost converter in the monitoring value obtained by the sensor, is the predicted voltage value of the RBF neural network DC microgrid prediction model, subscript i = 1, 2, ..., N, N is the number of boost converters, is the preset threshold for voltage network communication attacks, is the kth triggering moment of the ith boost converter for the voltage network communication attack, is the k+1th triggering moment of the ith boost converter for the voltage network communication attack, inf{·} is the infimum of the set, and t is a discrete time point.

[0050] Preferably, the super spiral sliding mode controller configured to converge the output value of the DC microgrid system according to the input parameter further comprises:

[0051] Based on the input parameters, calculating the average voltage and current coordination errors of any distributed generator set in the DC microgrid system;

[0052] Simulate network communication attacks and generate attack models;

[0053] Based on the attack model, determining the average voltage under attack and the coordinated error of the current under attack of any distributed generator set in the DC microgrid system under the network communication attack;

[0054] Based on the average voltage under the attack, calculating the average steady-state voltage of the DC microgrid system;

[0055] Analyze the dynamic change of the current coordination error under the attack to obtain a dynamic change model;

[0056] Substituting the steady-state voltage average value into the dynamic change model to obtain an updated dynamic change model;

[0057] Based on the updated dynamic change model, under the condition that the current cooperation error is zero, calculating the steady-state voltage deviation value of the DC microgrid system;

[0058] Substituting the steady-state voltage deviation value into the updated dynamic change model to obtain a steady-state current deviation value of the DC microgrid system;

[0059] Obtaining a first control variable of the boost converter according to a reference voltage of the boost converter, a voltage output by the boost converter, and a steady-state voltage deviation value;

[0060] Obtaining a second control variable of the boost converter according to a reference current of the current coordination error, a current output by the boost converter, and a steady-state current deviation value;

[0061] A super spiral sliding mode controller is set based on the first control variable, the second control variable, the sliding surface function, the reaching law and the nonlinear sigmoid function.

[0062] On the other hand, an embodiment of this specification provides a power grid system control device considering deterministic network communication attacks, the device comprising:

[0063] A control model building module is used to build a control model of a DC microgrid system;

[0064] A prediction model building module, used to build a RBF neural network DC microgrid prediction model based on the control model of the DC microgrid system;

[0065] A deviation value calculation module, used to calculate the deviation value between the monitoring value of the DC microgrid system obtained by the sensor monitoring and the predicted value of the RBF neural network DC microgrid prediction model;

[0066] A judgment module is used to judge whether a preset event trigger mechanism is triggered based on the deviation value and a preset threshold value, wherein the preset event trigger mechanism indicates that the DC microgrid system is subjected to a network communication attack; if so, the predicted value of the RBF neural network DC microgrid prediction model is used as an input parameter of the super spiral sliding mode controller; if not, the monitoring value is used as an input parameter of the super spiral sliding mode controller;

[0067] The setting module is used to set a super spiral sliding mode controller for converging the output value of the DC microgrid system according to the input parameters.

[0068] On the other hand, an embodiment of the present specification further provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein when the computer program is executed by the processor, the instructions of any one of the above methods are executed.

[0069] On the other hand, an embodiment of the present specification further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor of a computer device, the computer program executes instructions of any one of the above-described methods.

[0070] On the other hand, the embodiments of this specification further provide a computer program product, which, when executed by a processor of a computer device, executes instructions of any one of the above methods.

[0071] It can be seen from the technical solutions provided in the above embodiments of this specification that, by using the method of the embodiments of this specification, a control model of a DC microgrid system is constructed, and an RBF neural network is introduced for prediction, which can effectively improve the operation effect of the system. The deviation between the sensor monitoring value and the predicted value of the RBF neural network prediction model is used to determine whether to trigger a preset event mechanism, such as a network communication attack. If an attack is detected, the system will use the predicted value as the input of the super-helical sliding mode controller to ensure that the system can make reasonable adjustments in the face of anomalies to prevent performance degradation; if no attack is detected, the actual monitoring value is used for control to ensure that the system remains efficient under normal circumstances. The super-helical sliding mode controller can dynamically adjust the system output according to the input parameters so that it converges within the expected range, thereby improving the stability and response speed of the system. In general, this process not only improves the stability and robustness of the system, but also optimizes the control effect through real-time monitoring and adjustment, ensuring that the system can operate efficiently and stably under various working conditions.

[0072] In order to make the above and other purposes, features and advantages of the present specification more obvious and easy to understand, the following specifically cites preferred embodiments and describes them in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0074] Figure 1 A schematic flow chart of a power grid system control method considering deterministic network communication attacks provided in an embodiment of this specification is shown;

[0075] Figure 2 A schematic diagram of a flow chart of establishing a control model for a DC microgrid system provided in an embodiment of this specification is shown;

[0076] Figure 3 A schematic diagram of a process for establishing a RBF neural network DC microgrid prediction model provided in an embodiment of this specification is shown;

[0077] Figure 4 A schematic diagram of a process for optimizing an initial model provided in an embodiment of this specification is shown;

[0078] Figure 5 A schematic diagram of a process flow of a super spiral sliding mode controller provided in an embodiment of this specification and used to converge the output value of a DC microgrid system is shown;

[0079] Figure 6 A schematic diagram of the module structure of a power grid system control device considering deterministic network communication attacks provided in an embodiment of this specification is shown;

[0080] Figure 7 A schematic diagram of the structure of a computer device provided in an embodiment of this specification is shown.

[0081] Description of the accompanying symbols:

[0082] 100. Control model building module;

[0083] 200. Prediction model building module;

[0084] 300. Deviation value calculation module;

[0085] 400, judgment module;

[0086] 500, setting module;

[0087] 702. Computer equipment;

[0088] 704, processor;

[0089] 706. Memory;

[0090] 708, driving mechanism;

[0091] 710, input / output module;

[0092] 712. Input devices;

[0093] 714. Output devices;

[0094] 716. Presentation equipment;

[0095] 718. Graphical user interface;

[0096] 720, network interface;

[0097] 722. Communication link;

[0098] 724. Communication bus. DETAILED DESCRIPTION

[0099] The following will be combined with the drawings in the embodiments of this specification to clearly and completely describe the technical solutions in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the embodiments of this specification.

[0100] 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, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant countries and regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0101] As an emerging network communication architecture, deterministic networks rely on their core technologies of resource reservation, service guarantee and explicit routing to set high standards for data transmission performance. Under this framework, the information exchange between the controllers of the distributed controlled DC microgrid system is highly dependent on the communication network, so it is more vulnerable to network communication attacks, which may cause the microgrid system to deviate from the rated operating point or even cause power outages. In order to improve the stability of the system, a series of measures need to be taken to address these challenges and ensure that the power grid system operates safely and stably under network communication attacks.

[0102] In order to solve the above problems, an embodiment of this specification provides a power grid system control method considering deterministic network communication attacks. Figure 1 This is a flowchart of a power grid system control method considering deterministic network communication attacks provided by an embodiment of this specification. This specification provides method operation steps as described in the embodiment or flowchart, but may include more or fewer operation steps based on conventional or non-creative labor. The order of steps listed in the embodiment is only one way of executing the order of many steps and does not represent the only execution order. When the system or device product is executed in practice, it can be executed in the order of the method shown in the embodiment or the accompanying drawings or in parallel.

[0103] It should be noted that the terms "first", "second", etc. in the description and claims of the embodiments of this specification and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of this specification described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, device, product or equipment that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0104] Reference Figure 1, the embodiment of this specification provides a power grid system control method considering deterministic network communication attacks, including:

[0105] S101: Establish a control model for the DC microgrid system;

[0106] S102: Establishing a RBF neural network DC microgrid prediction model based on the control model of the DC microgrid system;

[0107] S103: Calculating a deviation between a monitoring value of the DC microgrid system obtained by monitoring by a sensor and a predicted value of the RBF neural network DC microgrid prediction model;

[0108] S104: Based on the deviation value and the preset threshold, determining whether a preset event triggering mechanism is triggered, wherein the preset event triggering mechanism indicates that the DC microgrid system is subject to a network communication attack;

[0109] S105: If yes, the predicted value of the RBF neural network DC microgrid prediction model is used as an input parameter of the super spiral sliding mode controller;

[0110] S106: If not, using the monitoring value as an input parameter of a super-helical sliding mode controller;

[0111] S107: According to the input parameters, a super spiral sliding mode controller is set to converge the output value of the DC microgrid system.

[0112] By using the method of the embodiment of this specification, a control model of a DC microgrid system is constructed, and an RBF neural network is introduced for prediction, which can effectively improve the operation effect of the system. The deviation between the sensor monitoring value and the predicted value of the RBF neural network prediction model is used to determine whether to trigger a preset event mechanism, such as a network communication attack. If an attack is detected, the system will use the predicted value as the input of the super-helical sliding mode controller to ensure that the system can make reasonable adjustments in the face of anomalies to prevent performance degradation; if no attack is detected, the actual monitoring value is used for control to ensure that the system remains efficient under normal circumstances. The super-helical sliding mode controller can dynamically adjust the system output according to the input parameters so that it converges within the expected range, thereby improving the stability and response speed of the system. In general, this process not only improves the stability and robustness of the system, but also optimizes the control effect through real-time monitoring and adjustment, ensuring that the system can operate efficiently and stably under various working conditions.

[0113] In order to prevent the deviation value between the sensor-based monitoring value and the predicted value of the RBF neural network DC microgrid prediction model for network communication attacks from exceeding a preset threshold during load changes, thereby causing false alarms of network communication attacks, the embodiment of this specification designs a nonlinear time-varying trigger function that incorporates nonlinear functions and network communication attack errors. When the nonlinear time-varying trigger function exceeds the preset threshold, the system is deemed to have suffered a network communication attack, and the predicted value of the RBF neural network DC microgrid prediction model replaces the sensor's monitoring value as the input of the super-spiral sliding mode controller, thereby effectively isolating the impact of network communication attacks on the power grid system.

[0114] By analyzing the steady-state voltage deviation and steady-state current deviation caused by voltage network communication attacks and current network communication attacks respectively, and further integrating the output voltage and output current of the boost converter to construct the sliding mode surface, a super spiral sliding mode controller is designed. At the same time, a smooth nonlinear sigmoid function is introduced, so that the actual operating state of the power grid system can converge to the target state quickly and smoothly, effectively curbing the impact of network communication attacks on the stability of the power grid system.

[0115] In the embodiments of this specification, refer to Figure 2 , the control model of establishing the DC microgrid system further includes:

[0116] S201: Setting up multiple distributed generator sets;

[0117] S202: Any distributed generator set is connected to a busbar through a boost converter, and each busbar is connected through a resistor and an inductor to form a DC microgrid system;

[0118] S203: Establish a mathematical model of the boost converter in one switching cycle as a control model of the DC microgrid system.

[0119] The mathematical model of the boost converter in one switching cycle includes:

[0120]

[0121]

[0122]

[0123] Among them, L i,f is the inductance of the i-th boost converter, C i,f is the capacitance of the i-th boost converter, For the i-th boost converter through the inductor L f The current, i i,o 、v i,oare the output current and output voltage of the i-th boost converter in the monitoring values ​​monitored by the sensor, d i is the duty cycle of the i-th boost converter switch, r i,d is the equivalent resistance of the i-th boost converter, V i,n is the output voltage of the distributed generator set corresponding to the i-th boost converter, subscript i=1, 2,…, N, N is the number of boost converters, and t is a discrete time point.

[0124] In the embodiments of this specification, refer to Figure 3 The control model of the DC microgrid system is based on the RBF neural network DC microgrid prediction model, which further includes:

[0125] S301: Establishing an RBF neural network DC microgrid voltage initial model and an RBF neural network DC microgrid current initial model respectively;

[0126] S302: Based on actual input of the mathematical model of the boost converter in one switching cycle under different load changes, the actual input is used as input data of the RBF neural network DC microgrid voltage initial model and the RBF neural network DC microgrid current initial model respectively;

[0127] S303: inputting the input data into the RBF neural network DC microgrid voltage initial model and the RBF neural network DC microgrid current initial model respectively to obtain output data of the two initial models;

[0128] S304: Based on the actual output of the mathematical model of the boost converter in one switching cycle under different load changes and the output data of the two initial models, the RBF neural network DC microgrid voltage initial model and the RBF neural network DC microgrid current initial model are optimized respectively to obtain the RBF neural network DC microgrid voltage prediction model and the RBF neural network DC microgrid current prediction model.

[0129] The actual input of the mathematical model of the boost converter in one switching cycle under different load changes includes duty cycle, inductance and current, etc. The actual input is used as the input data of the RBF neural network DC microgrid voltage initial model and the RBF neural network DC microgrid current initial model, respectively. The input data is:

[0130]

[0131] in, is the input data of the RBF neural network DC microgrid voltage initial model, or the input data of the RBF neural network DC microgrid current initial model, that is, the input characteristic data set of the i-th boost converter under different load changes; For the i-th boost converter through the inductor L f The current, d i is the duty cycle of the switch of the i-th boost converter, subscript i=1,2,…,N, N is the number of boost converters;

[0132] The output data of the RBF neural network DC microgrid voltage initial model and the RBF neural network DC microgrid current initial model are:

[0133] y i,k =w i,k,1 h i,k,1 +w i,k,2 h i,k,2 +…+w i,k,j h i,k,j +…+w i,k,l h i,k,l (5)

[0134]

[0135] Among them, when k = 1, y i,k is the output data of the initial model of DC microgrid voltage using RBF neural network, w i,k,j is the weight value of the jth hidden layer neuron in the RBF neural network DC microgrid voltage initial model, h i,k,j is the output of the jth hidden layer neuron in the RBF neural network DC microgrid voltage initial model, b i,k,j is the width of the Gaussian basis function of the jth hidden layer neuron in the RBF neural network DC microgrid voltage initial model, and b i,k,j >0,c i,k,j =[c i,k,j,1 c i,k,j,2 ] T , c i,k,j is the center point vector value of the jth hidden layer neuron, c i,k,j,m is the mth component of the center point vector value of the jth hidden layer neuron, m = 1, 2, j = 1, 2, ..., l, l represents the number of hidden layer neurons;

[0136] When k = 2, y i,k is the output data of the initial current model of the RBF neural network DC microgrid, w i,k,j is the weight value of the jth hidden layer neuron in the RBF neural network DC microgrid current initial model, h i,k,jis the output of the jth hidden layer neuron in the RBF neural network DC microgrid current initial model, b i,k,j is the width of the Gaussian basis function of the jth hidden layer neuron in the RBF neural network DC microgrid current initial model, and b i,k,j >0,c i,k,j =[c i,k,j,1 c i,k,j,2 ] T , c i,k,j is the center point vector value of the jth hidden layer neuron, c i,k,j,m is the mth component of the center point vector value of the jth hidden layer neuron, m = 1, 2, j = 1, 2, ..., l, l represents the number of hidden layer neurons;

[0137] Reference Figure 4 , based on the actual output of the mathematical model of the boost converter in one switching cycle under different load changes and the output data of the two initial models, the RBF neural network DC microgrid voltage initial model and the RBF neural network DC microgrid current initial model are optimized respectively to obtain the RBF neural network DC microgrid voltage prediction model and the RBF neural network DC microgrid current prediction model further include:

[0138] S401: calculating an error value between the actual output and the output data based on the actual output of the mathematical model of the boost converter in one switching cycle under different load changes and the output data of the two initial models;

[0139] S402: Adopting the Adam gradient descent method to optimize the model parameters of the two initial models respectively, to obtain the RBF neural network DC microgrid voltage prediction model and the RBF neural network DC microgrid current prediction model corresponding to the minimized error value.

[0140] The actual output of the mathematical model of the boost converter in one switching cycle under different load changes includes current and voltage.

[0141] The calculating the error value between the actual output and the output data based on the actual output of the mathematical model of the boost converter in one switching cycle under different load changes and the output data of the two initial models further comprises:

[0142] The error value between the actual output and the output data is calculated by the following formula:

[0143]

[0144] Among them, when k = 1, y i,k is the output data of the initial model of RBF neural network DC microgrid voltage, is the actual output voltage in the actual output, E i,k is the error value between the actual output voltage and the output data of the RBF neural network DC microgrid voltage initial model; when k = 2, y i,k is the output data of the initial model of the DC microgrid current using the RBF neural network, is the actual output current in the actual output, E i,k is the error value between the actual output current and the output data of the RBF neural network DC microgrid current initial model.

[0145] The Adam gradient descent method is used to optimize the model parameters of the two initial models respectively, and the weight value w i,k,j , the center point c of the hidden layer neuron i,k,j,m and the width b of the Gaussian basis function i,k,j The gradient of is:

[0146]

[0147] Among them, Δw i,k,j is the gradient of the weight value, Δb i,k,j is the gradient of the width of the Gaussian basis function, Δc i,k,j,m is the gradient of the center point of the hidden layer neuron. When m=1, When m = 2

[0148] When k = 1, w i,k,j is the weight value of the jth hidden layer neuron in the RBF neural network DC microgrid voltage initial model, is the input data of the initial voltage model of the RBF neural network DC microgrid, y i,k is the output data of the initial model of RBF neural network DC microgrid voltage, is the actual output voltage in the actual output, E i,k is the error value between the actual output voltage and the output data of the RBF neural network DC microgrid voltage initial model, h i,k,j is the output of the jth hidden layer neuron in the RBF neural network DC microgrid voltage initial model;

[0149] When k = 2, w i,k,j is the weight value of the jth hidden layer neuron in the RBF neural network DC microgrid current initial model, is the input data of the initial model of the DC microgrid current of the RBF neural network, y i,k is the output data of the initial model of the DC microgrid current using the RBF neural network, is the actual output current in the actual output, E i,kis the error value between the actual output current and the output data of the RBF neural network DC microgrid current initial model, h i,k,j is the output of the jth hidden layer neuron in the initial current model of the RBF neural network DC microgrid;

[0150] b i,k,j is the width of the Gaussian basis function of the jth hidden layer neuron.

[0151] According to formula (8), the Adam gradient descent method is used as the parameter w i,k,j Design an adaptive learning rate to update it:

[0152]

[0153] Among them, m a is the initial model in the current cycle w i,k,j The first moment estimate of the gradient, m a t-T is the initial model in the previous cycle w i,k,j The first moment estimate of the gradient, v a w i,k,j The initial model's second-order moment estimate of the gradient in the current cycle, v a t-T is the initial model in the previous cycle w i,k,j The second moment estimate of the gradient, is the first-order moment estimate of the error correction, is the decay rate of the first-order moment estimate, is the decay rate of the second-order moment estimate, is the error-corrected second-order moment estimate, w i,k,j is the weight value of the jth hidden layer neuron in the initial model in the current cycle, w i，k，j t-T is the weight value of the jth hidden layer neuron in the previous cycle of the initial model, η a w i,k,j The adaptive learning rate, Λ a , β a1 , β a2 ,∈ a is a constant.

[0154] According to formula (8), the Adam gradient descent method is used as the parameter b i,k,j Design an adaptive learning rate to update it:

[0155]

[0156] Among them, m b b is the initial model in the current cycle i,k,jThe first moment estimate of the gradient, m b t-T b is the initial model in the previous cycle i,k,j The first moment estimate of the gradient, v b b is the initial model in the current cycle i,k,j The second moment estimate of the gradient, v b t-T b is the initial model in the previous cycle i,k,j The second moment estimate of the gradient, is the first-order moment estimate of the error correction, is the error-corrected second-order moment estimate, is the decay rate of the first-order moment estimate, is the decay rate of the second-order moment estimate, b i,k,j is the width of the Gaussian basis function of the jth hidden layer neuron in the initial model in the current cycle, b i，k，j t-T is the width of the Gaussian basis function of the jth hidden layer neuron in the initial model in the previous cycle, η b for b i,k,j The adaptive learning rate, Λ b , β b1 , β b2 ,∈ b is a constant;

[0157] According to formula (8), the Adam gradient descent method is used as the parameter c i,k,j,m Design an adaptive learning rate to update it:

[0158]

[0159] Among them, m c is the initial model in the current cycle c i,k,j,m The first moment estimate of the gradient, m c t-T is the initial model in the previous cycle c i,k,j,m The first moment estimate of the gradient, v c is the initial model in the current cycle c i,k,j,m The second moment estimate of the gradient, v c t-T is the initial model in the previous cycle c i,k,j,m The second moment estimate of the gradient, is the first-order moment estimate of the error correction, is the error-corrected second-order moment estimate, is the decay rate of the first-order moment estimate, is the decay rate of the second-order moment estimate, η c is parameter c i,k,j,mThe adaptive learning rate, Λ c , β c1 , β c2 ,∈ c is a constant, c i,k,j,m is the mth component of the center point vector value of the jth hidden layer neuron in the current cycle, c i，k，j，m t-T is the mth component of the center point vector value of the jth hidden layer neuron in the previous cycle, where m is 1 or 2.

[0160] In the embodiment of the present specification, the deviation value between the monitoring value of the DC microgrid system obtained by the sensor monitoring and the predicted value of the RBF neural network DC microgrid prediction model is calculated.

[0161] Specific:

[0162]

[0163] Among them, e i,I is the deviation value corresponding to the current when the current network communication is attacked, i i,o is the current output by the i-th boost converter in the monitoring value monitored by the sensor, It is the predicted current value of the RBF neural network DC microgrid prediction model.

[0164]

[0165] Among them, e i,V is the voltage deviation value corresponding to the voltage network communication attack, v i,o is the voltage output by the i-th boost converter in the monitoring value obtained by the sensor, It is the predicted voltage value of the RBF neural network DC microgrid prediction model.

[0166] Based on the deviation value and the preset threshold, it is determined whether the preset event trigger mechanism is triggered, and the preset event trigger mechanism indicates that the DC microgrid system is under network communication attack.

[0167] The preset event trigger mechanism includes: a preset event trigger mechanism for current network communication attacks, and a preset event trigger mechanism for voltage network communication attacks;

[0168] The preset event trigger mechanism for current network communication attacks is characterized by the following formula:

[0169]

[0170] Among them, Θ i,I is the nonlinear time-varying trigger function for current network communication attacks, κ i,1 , κi,2 , κ i,3 , κ i,4 , are constants, e i,I is the deviation value corresponding to the current during the current network communication attack, i i,o is the current output by the i-th boost converter in the monitoring value monitored by the sensor, is the predicted current value of the RBF neural network DC microgrid prediction model, that is, y i,2 , subscript i = 1, 2, ..., N, N is the number of boost converters, is the preset threshold for current network communication attacks, is the kth triggering moment of the ith boost converter for the current network communication attack, is the k+1th triggering moment of the ith boost converter for the current network communication attack, inf{·} is the infimum of the set, and t is a discrete time point;

[0171] The preset event trigger mechanism for voltage network communication attacks is characterized by the following formula:

[0172]

[0173] Among them, Θ i,V is the nonlinear time-varying trigger function for voltage network communication attacks, κ i,5 , κ i,6 , κ i,7 , κ i,8 , is a constant, e i,V is the voltage deviation value corresponding to the voltage network communication attack, v i,o is the voltage output by the i-th boost converter in the monitoring value obtained by the sensor, is the predicted voltage value of the RBF neural network DC microgrid prediction model, that is, y i,1 , subscript i = 1, 2, ..., N, N is the number of boost converters, is the preset threshold for voltage network communication attacks, is the kth triggering moment of the ith boost converter for the voltage network communication attack, is the k+1th triggering moment of the ith boost converter for the voltage network communication attack, inf{·} is the infimum of the set, and t is a discrete time point.

[0174] If the preset event trigger mechanism is triggered, the predicted value of the RBF neural network DC microgrid prediction model is used as the input parameter of the super spiral sliding mode controller:

[0175]

[0176]

[0177] in, is the current parameter in the input parameters, is the voltage parameter in the input parameters.

[0178] If the preset event trigger mechanism is not triggered, the monitoring value is used as the input parameter of the super-helical sliding mode controller:

[0179]

[0180]

[0181] in, is the current parameter in the input parameters, is the voltage parameter in the input parameters.

[0182] In the embodiments of this specification, refer to Figure 5 , the super spiral sliding mode controller configured to converge the output value of the DC microgrid system according to the input parameters further comprises:

[0183] S501: Calculating average voltage and current coordination errors of any distributed generator set in the DC microgrid system based on the input parameters;

[0184] S502: Simulate network communication attacks and generate attack models;

[0185] S503: Based on the attack model, determining the average voltage under attack and the coordinated error of the current under attack of any distributed generator set in the DC microgrid system under the network communication attack;

[0186] S504: Calculating a steady-state voltage average value of the DC microgrid system based on the average voltage under the attack;

[0187] S505: Analyze the dynamic change of the current coordination error under the attack to obtain a dynamic change model;

[0188] S506: Substituting the steady-state voltage average value into the dynamic change model to obtain an updated dynamic change model;

[0189] S507: Based on the updated dynamic change model, under the condition that the current cooperation error is zero, calculating the steady-state voltage deviation value of the DC microgrid system;

[0190] S508: Substituting the steady-state voltage deviation value into the updated dynamic change model to obtain a steady-state current deviation value of the DC microgrid system;

[0191] S509: Obtaining a first control variable of the boost converter according to a reference voltage of the boost converter, a voltage output by the boost converter, and a steady-state voltage deviation value;

[0192] S510: Obtaining a second control variable of the boost converter according to a reference current of the current coordination error, a current output by the boost converter, and a steady-state current deviation value;

[0193] S511: Based on the first control variable, the second control variable, the sliding surface function, the reaching law and the nonlinear sigmoid function, a super spiral sliding mode controller is set.

[0194] Based on the input parameters, the average voltage and current coordination error of any distributed generator set in the DC microgrid system is calculated as:

[0195]

[0196]

[0197] in, is the average voltage of the i-th distributed generator set, is the average voltage of the jth distributed generator set, is the voltage parameter of the i-th distributed generator set in the input parameters, a ij is the information exchange value between the i-th distributed generator set and the j-th distributed generator set, if there is information exchange, it is 1, if there is no information exchange, it is 0, N is the number of boost converters;

[0198] ω i is the current coordination error of the i-th distributed generator set, is the current parameter of the i-th distributed generator set in the input parameters, is the current parameter of the jth distributed generator set in the input parameters.

[0199] Rewriting formulas (20) and (21) into matrix form is:

[0200]

[0201] W i =-LI o (twenty three)

[0202] in,

[0203]

[0204]

[0205] , W i =[ω1 ω2 … ω N ] T .

[0206] The reference voltage of the i-th boost converter is further obtained as:

[0207]

[0208] in, is the reference voltage of the i-th boost converter, is the voltage correction term, is the current correction term, are the proportional coefficient and integral coefficient of the voltage controller respectively, are the proportional coefficient and integral coefficient of the current controller respectively, s is the Laplace operator, is the average voltage of the i-th distributed generator set, ω i is the current coordination error of the i-th distributed generator set, is the reference current of the current coordination error, generally 0, v nom is the nominal voltage value, is the current parameter of the i-th distributed generator set in the input parameters, r i,d is the equivalent resistance of the i-th boost converter, and N is the number of boost converters.

[0209] Furthermore, the network communication attack is simulated and the attack model is generated as follows:

[0210]

[0211] in, is the attack data value, xi is the monitoring value of the DC microgrid system obtained by the sensor, and the monitoring value includes voltage or current, is the amplitude of false data, δ i Indicates whether there is an attack. If there is an attack, then δ i is 1, if there is no attack, then δ i is 0.

[0212] Based on the attack model formulas (22) and (23), the average voltage and current coordination error under attack of any distributed generator set in the DC microgrid system under network communication attack are determined as follows:

[0213]

[0214]

[0215] in, is the average voltage under attack, is the current coordination error under attack,

[0216] α=[α1 α2 … α N ] T , β=[β1 β2 … β N ] T ,

[0217]

[0218]

[0219] α i Indicates whether there is a voltage network communication attack. If there is an attack, then α i is 1, if there is no attack, then α i is 0, β i Indicates whether there is a current network communication attack. If there is an attack, β i is 1, if there is no attack, then β i is 0, The amplitude of the false data injected for voltage network communication attack, The amplitude of injecting false data for current network communication attacks, is the final average voltage value of the current network communication attack, is the final current coordination error value after voltage network communication attack, i=1,2,…,N, N is the number of boost converters.

[0220] Based on the formula (25) of the average voltage under attack, the average steady-state voltage of the DC microgrid system is calculated as:

[0221]

[0222] in, is the average steady-state voltage, is the average voltage value when there is no network communication attack.

[0223]

[0224] Derivative formula (26) and combined with formulas (20), (21), (22), and (23), in steady state, All are 0. The dynamic change of current coordination error under attack is analyzed, and the dynamic change model is obtained:

[0225]

[0226] Among them, V nom =[v nom ,v nom ,…,v nom ] T ∈R n×1 ,

[0227] The steady-state voltage average Substituting into the dynamic change model formula (28), the updated dynamic change model is obtained:

[0228]

[0229] Based on the updated dynamic change model formula (29), under the condition that the current cooperation error is zero, that is, Calculate the steady-state voltage deviation value of the DC microgrid system:

[0230]

[0231] Among them, v is the steady-state voltage deviation value of the DC microgrid system,

[0232]

[0233] N is the number of boost converters.

[0234] Substituting the steady-state voltage deviation value formula (30) into the updated dynamic change model formula (29), the steady-state current deviation value of the DC microgrid system is obtained:

[0235]

[0236] Among them, I is the steady-state current deviation value of the DC microgrid system.

[0237] According to the reference voltage of the boost converter, the voltage output by the boost converter and the steady-state voltage deviation value, the first control variable of the boost converter is obtained:

[0238]

[0239] Among them, x i,1 is the first control variable of the i-th boost converter, is the reference voltage of the i-th boost converter, v i,o is the voltage output by the i-th boost converter in the monitoring value obtained by the sensor, i,V is the steady-state voltage deviation value of the i-th boost converter, is the predicted voltage value of the RBF neural network DC microgrid prediction model, and N is the number of boost converters.

[0240] According to the reference current of the current coordination error, the current output by the boost converter and the steady-state current deviation value, the second control variable of the boost converter is obtained:

[0241]

[0242] Among them, x i,2 is the second control variable of the i-th boost converter, is the reference current of the current coordination error, i i,o is the current output by the i-th boost converter in the monitoring value obtained by the sensor, i,I is the steady-state current deviation value of the i-th boost converter, is the current parameter in the input parameters, It is the predicted current value of the RBF neural network DC microgrid prediction model.

[0243] In the embodiment of this specification, the sliding surface function s i for:

[0244] s i =c i,1 x i,1 +c i,2 x i,2 (34)

[0245] Among them, s i is the sliding surface function, c i,1 、c i,2 is a constant.

[0246] Reaching Law i for:

[0247]

[0248] Among them, u i is the reaching law, sign(g) is the sign function, λ i,1 , i,2 is a constant and satisfies λ i,1 >0,λ i,2 >0.

[0249] In order to reduce the chattering phenomenon that may occur in the DC microgrid system under the control of the super-helical sliding mode controller, a smoother nonlinear sigmoid function is introduced, which can effectively alleviate the sudden changes and high-frequency oscillations in the control signal while maintaining the control accuracy, significantly reduce the occurrence of chattering, and improve the stability and control performance of the system.

[0250] Based on the following elements: the first control variable and the second control variable are used to adjust the voltage and current to maintain the stability of the system, the sliding surface function is used to define the target position of the system state, and the reaching law and the nonlinear Sigmoid function are used to improve the control performance to ensure that the system can remain stable in the presence of disturbances. A super helical sliding mode controller is thus obtained.

[0251] This application provides users with corresponding big data analysis (such as personal biometrics, identity data, consumption data, asset data, electronic terminal operation data, etc.) operation entrances for users to choose to agree or reject automated decision-making results; if the user chooses to reject, the expert decision-making process will be entered.

[0252] Based on the above-mentioned power grid system control method taking into account deterministic network communication attacks, the embodiments of this specification also provide a power grid system control device taking into account deterministic network communication attacks.

[0253] The device may include a system (including a distributed system), software (application), module, component, server, client, etc. that uses the method described in the embodiments of this specification and is combined with necessary implementation hardware. Based on the same innovative concept, the device in one or more embodiments provided in the embodiments of this specification is described in the following embodiments.

[0254] Since the implementation scheme and method of the device to solve the problem are similar, the implementation of the specific device in the embodiment of this specification can refer to the implementation of the aforementioned method, and the repetitions will not be repeated. As used below, the term "unit" or "module" can implement a combination of software and / or hardware for a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.

[0255] Specifically, Figure 6 This is a schematic diagram of the module structure of an embodiment of a power grid system control device considering deterministic network communication attacks provided by the embodiments of this specification, with reference to Figure 6 As shown, an embodiment of the present specification provides a power grid system control device that takes into account deterministic network communication attacks, including: a control model building module 100, a prediction model building module 200, a deviation value calculation module 300, a judgment module 400, and a setting module 500.

[0256] A control model building module 100 is used to build a control model of a DC microgrid system;

[0257] A prediction model building module 200 is used to build a RBF neural network DC microgrid prediction model based on the control model of the DC microgrid system;

[0258] The deviation value calculation module 300 is used to calculate the deviation value between the monitoring value of the DC microgrid system obtained by the sensor monitoring and the predicted value of the RBF neural network DC microgrid prediction model;

[0259] The judgment module 400 is used to judge whether a preset event trigger mechanism is triggered based on the deviation value and the preset threshold value, and the preset event trigger mechanism indicates that the DC microgrid system is subjected to a network communication attack; if so, the predicted value of the RBF neural network DC microgrid prediction model is used as an input parameter of the super spiral sliding mode controller; if not, the monitoring value is used as an input parameter of the super spiral sliding mode controller;

[0260] The setting module 500 is used to set a super spiral sliding mode controller for converging the output value of the DC microgrid system according to the input parameters.

[0261] Reference Figure 7 As shown, based on the above-mentioned power grid system control method considering deterministic network communication attacks, an embodiment of this specification further provides a computer device 702, wherein the above-mentioned method runs on the computer device 702.

[0262] The computer device 702 may include one or more processors 704, such as one or more central processing units (CPUs) or graphics processing units (GPUs), each of which may implement one or more hardware threads. The computer device 702 may also include any memory 706, which is used to store any kind of information such as code, settings, data, etc. In a specific embodiment, a computer program on the memory 706 and executable on the processor 704, when the computer program is executed by the processor 704, can execute instructions according to the above method. Non-limiting, for example, the memory 706 may include any one or more combinations of the following: any type of RAM, any type of ROM, flash memory device, hard disk, optical disk, etc. More generally, any memory may use any technology to store information. Further, any memory may provide volatile or non-volatile retention of information. Further, any memory may represent a fixed or removable component of the computer device 702. In one case, when the processor 704 executes an associated instruction stored in any memory or combination of memories, the computer device 702 may perform any operation of the associated instruction. The computer device 702 also includes one or more drive mechanisms 708 for interacting with any storage, such as a hard disk drive mechanism, an optical disk drive mechanism, etc.

[0263] The computer device 702 may also include an input / output module 710 (I / O) for receiving various inputs (via input devices 712) and for providing various outputs (via output devices 714). A specific output mechanism may include a presentation device 716 and an associated graphical user interface 718 (GUI). In other embodiments, the input / output module 710 (I / O), input device 712, and output device 714 may not be included, and the computer device 702 may be used as a computer device in a network. The computer device 702 may also include one or more network interfaces 720 for exchanging data with other devices via one or more communication links 722. One or more communication buses 724 couple the components described above together.

[0264] The communication link 722 may be implemented in any manner, for example, through a local area network, a wide area network (e.g., the Internet), a point-to-point connection, etc., or any combination thereof. The communication link 722 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers, etc. governed by any protocol or combination of protocols.

[0265] Corresponds to Figure 1-Figure 5 The method in the embodiment of the present specification also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above method are executed.

[0266] The embodiment of the present specification also provides a computer-readable instruction, wherein when the processor executes the instruction, the program therein causes the processor to execute the following Figures 1 to 5 The method shown.

[0267] The present specification also provides a computer program product. When the computer program product is executed by a processor of a computer device, the computer program product performs the following steps: Figures 1 to 5 The method shown.

[0268] The computer program product described in this specification is a software product that mainly implements the method described in this specification through a computer program.

[0269] It should be understood that in the various embodiments of this specification, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this specification.

[0270] It should also be understood that in the embodiments of this specification, the term "and / or" is only a description of the association relationship of the associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in the embodiments of this specification generally indicates that the associated objects before and after are in an "or" relationship.

[0271] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the embodiments of this specification can be implemented with electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the embodiments of this specification.

[0272] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0273] In the several embodiments provided in this specification, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, or it can be an electrical, mechanical or other form of connection.

[0274] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of this specification.

[0275] In addition, each functional unit in each embodiment of this specification may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0276] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium.

[0277] Based on this understanding, the technical solution of the embodiments of this specification is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of this specification. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), disk or optical disk and other media that can store program codes.

[0278] Specific embodiments are used in this specification to illustrate the principles and implementation methods of the embodiments of this specification. The description of the above embodiments is only used to help understand the methods and core ideas of the embodiments of this specification. At the same time, for those skilled in the art, according to the ideas of the embodiments of this specification, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on the embodiments of this specification.

Claims

1. A power grid system control method considering deterministic network communication attacks, characterized in that: include: Establish the control model of DC microgrid system; Based on the control model of the DC microgrid system, a RBF neural network DC microgrid prediction model is established; Calculate the deviation between the monitoring value of the DC microgrid system obtained by the sensor monitoring and the predicted value of the RBF neural network DC microgrid prediction model; Based on the deviation value and the preset threshold, determining whether a preset event trigger mechanism is triggered, wherein the preset event trigger mechanism indicates that the DC microgrid system is subject to a network communication attack; If yes, the predicted value of the RBF neural network DC microgrid prediction model is used as an input parameter of the super spiral sliding mode controller; If not, the monitoring value is used as an input parameter of the super-helical sliding mode controller; According to the input parameters, a super spiral sliding mode controller is set to converge the output value of the DC microgrid system.

2. The method according to claim 1, characterized in that The control model of establishing the DC microgrid system further includes: Set up multiple distributed generator sets; Any distributed generator set is connected to a busbar through a boost converter, and each busbar is connected through resistance and inductance to form a DC microgrid system; A mathematical model of the boost converter in one switching cycle is established as the control model of the DC microgrid system.

3. The method according to claim 2, characterized in that The mathematical model of the boost converter in one switching cycle includes: Among them, L i,f is the inductance of the i-th boost converter, C i,f is the capacitance of the i-th boost converter, For the i-th boost converter through the inductor L f The current, i i,o 、v i,o are the output current and output voltage of the i-th boost converter in the monitoring values ​​monitored by the sensor, d i is the duty cycle of the i-th boost converter switch, r i,d is the equivalent resistance of the i-th boost converter, V i,n is the output voltage of the distributed generator set corresponding to the i-th boost converter, subscript i=1, 2,…, N, N is the number of boost converters, and t is a discrete time point.

4. The method according to claim 2, characterized in that: The control model based on the DC microgrid system, establishing the RBF neural network DC microgrid prediction model further includes: Establish the RBF neural network DC microgrid voltage initial model and the RBF neural network DC microgrid current initial model respectively; Based on the actual input of the mathematical model of the boost converter in one switching cycle under different load changes, the input data are respectively used as the RBF neural network DC microgrid voltage initial model and the RBF neural network DC microgrid current initial model; Inputting the input data into the RBF neural network DC microgrid voltage initial model and the RBF neural network DC microgrid current initial model respectively to obtain output data of the two initial models; Based on the actual output of the mathematical model of the boost converter in one switching cycle under different load changes and the output data of the two initial models, the RBF neural network DC microgrid voltage initial model and the RBF neural network DC microgrid current initial model are optimized respectively to obtain the RBF neural network DC microgrid voltage prediction model and the RBF neural network DC microgrid current prediction model.

5. The method according to claim 4, characterized in that The input data is: in, is the input data of the RBF neural network DC microgrid voltage initial model, or the input data of the RBF neural network DC microgrid current initial model, For the i-th boost converter through the inductor L f The current, d i is the duty cycle of the switch of the i-th boost converter, subscript i=1,2,…,N, N is the number of boost converters; The output data is: y i,k =w i,k,1 h i,k,1 +w i,k,2 h i,k,2 +…+w i,k,j h i,k,j +…+w i,k,l h i,k,l ; When k = 1, y i,k is the output data of the initial model of the DC microgrid voltage using the RBF neural network, w i,k,j is the weight value of the jth hidden layer neuron in the RBF neural network DC microgrid voltage initial model, h i,k,j is the output of the jth hidden layer neuron in the RBF neural network DC microgrid voltage initial model, b i,k,j is the width of the Gaussian basis function of the jth hidden layer neuron in the RBF neural network DC microgrid voltage initial model, and b i,k,j >0,c i,k,j =[c i,k,j,1 c i,k,j,2 ] T , c i,k,j is the center point vector value of the jth hidden layer neuron, c i,k,j,m is the mth component of the center point vector value of the jth hidden layer neuron, m = 1, 2, j = 1, 2, ..., l, l represents the number of hidden layer neurons; When k = 2, y i,k is the output data of the initial current model of the RBF neural network DC microgrid, w i,k,j is the weight value of the jth hidden layer neuron in the RBF neural network DC microgrid current initial model, h i,k,j is the output of the jth hidden layer neuron in the RBF neural network DC microgrid current initial model, b i,k,j is the width of the Gaussian basis function of the jth hidden layer neuron in the RBF neural network DC microgrid current initial model, and b i,k,j >0,c i,k,j =[c i,k,j,1 c i,k,j,2 ] T , c i,k,j is the center point vector value of the jth hidden layer neuron, c i,k,j,m is the mth component of the center point vector value of the jth hidden layer neuron, m=1, 2, j=1, 2, …, l, l represents the number of hidden layer neurons.

6. The method according to claim 5, characterized in that The actual output of the mathematical model of the boost converter in a switching cycle under different load changes and the output data of the two initial models are used to optimize the RBF neural network DC microgrid voltage initial model and the RBF neural network DC microgrid current initial model respectively, and the RBF neural network DC microgrid voltage prediction model and the RBF neural network DC microgrid current prediction model are obtained, which further includes: Based on the actual output of the mathematical model of the boost converter in one switching cycle under different load changes and the output data of the two initial models, calculating an error value between the actual output and the output data; The Adam gradient descent method is used to optimize the model parameters of the two initial models respectively, and the RBF neural network DC microgrid voltage prediction model and the RBF neural network DC microgrid current prediction model corresponding to the minimized error value are obtained.

7. The method according to claim 6, characterized in that The calculating the error value between the actual output and the output data based on the actual output of the mathematical model of the boost converter in one switching cycle under different load changes and the output data of the two initial models further comprises: The error value between the actual output and the output data is calculated by the following formula: When k = 1, y i,k is the output data of the initial model of RBF neural network DC microgrid voltage, is the actual output voltage in the actual output, E i,k is the error value between the actual output voltage and the output data of the RBF neural network DC microgrid voltage initial model; when k = 2, y i,k is the output data of the initial model of the DC microgrid current using the RBF neural network, is the actual output current in the actual output, E i,k is the error value between the actual output current and the output data of the RBF neural network DC microgrid current initial model.

8. The method according to claim 1, characterized in that The preset event trigger mechanism includes: a preset event trigger mechanism for current network communication attacks, and a preset event trigger mechanism for voltage network communication attacks; The preset event trigger mechanism for current network communication attacks is characterized by the following formula: Among them, Θ i,I is the nonlinear time-varying trigger function for current network communication attacks, κ i,1 , κ i,2 , κ i,3 , κ i,4 , are constants, e i,I is the deviation value corresponding to the current when the current network communication is attacked, i i,o is the current output by the i-th boost converter in the monitoring value monitored by the sensor, is the predicted current value of the RBF neural network DC microgrid prediction model, subscript i = 1, 2, ..., N, N is the number of boost converters, is the preset threshold for current network communication attacks, is the kth triggering moment of the ith boost converter for the current network communication attack, is the k+1th triggering moment of the ith boost converter for the current network communication attack, inf{·} is the infimum of the set, and t is a discrete time point; The preset event trigger mechanism for voltage network communication attacks is characterized by the following formula: Among them, Θ i,V is the nonlinear time-varying trigger function for voltage network communication attacks, κ i,5 , κ i,6 , κ i,7 , κ i,8 , is a constant, e i,V is the voltage deviation value corresponding to the voltage network communication attack, v i,o is the voltage output by the i-th boost converter in the monitoring value obtained by the sensor, is the predicted voltage value of the RBF neural network DC microgrid prediction model, subscript i = 1, 2, ..., N, N is the number of boost converters, is the preset threshold for voltage network communication attacks, is the kth triggering moment of the ith boost converter for the voltage network communication attack, is the k+1th triggering moment of the ith boost converter for the voltage network communication attack, inf{·} is the infimum of the set, and t is a discrete time point.

9. The method according to claim 1, characterized in that: The super spiral sliding mode controller configured to converge the output value of the DC microgrid system according to the input parameters further comprises: Based on the input parameters, calculating the average voltage and current coordination errors of any distributed generator set in the DC microgrid system; Simulate network communication attacks and generate attack models; Based on the attack model, determining the average voltage under attack and the coordinated error of the current under attack of any distributed generator set in the DC microgrid system under the network communication attack; Based on the average voltage under the attack, calculating the average steady-state voltage of the DC microgrid system; Analyze the dynamic change of the current coordination error under the attack to obtain a dynamic change model; Substituting the steady-state voltage average value into the dynamic change model to obtain an updated dynamic change model; Based on the updated dynamic change model, under the condition that the current cooperation error is zero, calculating the steady-state voltage deviation value of the DC microgrid system; Substituting the steady-state voltage deviation value into the updated dynamic change model to obtain a steady-state current deviation value of the DC microgrid system; Obtaining a first control variable of the boost converter according to a reference voltage of the boost converter, a voltage output by the boost converter, and a steady-state voltage deviation value; Obtaining a second control variable of the boost converter according to a reference current of the current coordination error, a current output by the boost converter, and a steady-state current deviation value; A super spiral sliding mode controller is set based on the first control variable, the second control variable, the sliding surface function, the reaching law and the nonlinear sigmoid function.

10. A power grid system control device considering deterministic network communication attacks, characterized in that: The device comprises: A control model building module is used to build a control model of a DC microgrid system; A prediction model building module, used to build a RBF neural network DC microgrid prediction model based on the control model of the DC microgrid system; A deviation value calculation module, used to calculate the deviation value between the monitoring value of the DC microgrid system obtained by the sensor monitoring and the predicted value of the RBF neural network DC microgrid prediction model; A judgment module is used to judge whether a preset event trigger mechanism is triggered based on the deviation value and a preset threshold value, wherein the preset event trigger mechanism indicates that the DC microgrid system is subjected to a network communication attack; if so, the predicted value of the RBF neural network DC microgrid prediction model is used as an input parameter of the super spiral sliding mode controller; if not, the monitoring value is used as an input parameter of the super spiral sliding mode controller; The setting module is used to set a super spiral sliding mode controller for converging the output value of the DC microgrid system according to the input parameters.

11. A computer device comprising a memory, a processor, and a computer program stored in the memory, characterized in that: When the computer program is executed by the processor, the computer program executes the instructions of the method according to any one of claims 1 to 9.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor of a computer device, the computer program executes the instructions of the method according to any one of claims 1 to 9.

13. A computer program product, characterized in that When the computer program product is executed by a processor of a computer device, the computer program product executes the instructions of the method according to any one of claims 1 to 9.

Images