Power distribution network information system risk assessment method and system based on graph convolutional network

Abstract

The invention discloses a power distribution network information system risk assessment method and system based on a graph convolutional network, and relates to the technical field of power distribution network information physical systems, and the method comprises the steps: capturing the time dynamic characteristics in a fault propagation process through combining with an LSTM, and precisely assessing the risk level of node voltage out-of-limit; a node voltage out-of-limit condition is taken as a core evaluation standard, and risk distribution in different attack scenes is comprehensively reflected; electrical characteristics of the system are learned under the non-fault condition, the evaluation precision is improved, and power distribution network information system risk evaluation is completed. According to the method, the combined model based on the GCN and the LSTM is introduced, the space-time characteristics in the power distribution network information physical system can be effectively captured, the node voltage out-of-limit risk level can be accurately evaluated, the risk evaluation precision and real-time performance can be remarkably improved, high-risk nodes can be identified, and the risk evaluation accuracy and real-time performance can be improved. And reliable support can be provided for safe operation of the power grid, so that the safety and toughness of the power distribution network are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of information-physical systems for distribution networks, and in particular to a risk assessment method and system for distribution network information systems based on graph convolutional networks. Background Art

[0002] At present, with the rapid development of distributed renewable energy and the in-depth application of information and communication technology, the information-physical system of distribution network has become an important part of modern smart grid. In CPDS, the information system is highly coupled with the physical power system, which not only realizes real-time monitoring and automatic control, but also makes the system face more and more information security threats, such as attacks on the control command downlink communication link. Such attacks may cause controllable resources to go offline, further cause node voltage to exceed the limit, and seriously threaten the safe operation of the distribution network.

[0003] At present, the risk assessment methods of distribution networks are mainly focused on static analysis and single scenario simulation, lacking a comprehensive capture of the dynamic characteristics of the system. Traditional assessment methods rely on a static process of gradual repair after a fault, which makes it difficult to reflect the spatiotemporal correlation of information attacks on the operating status of the distribution network. This method has significant deficiencies in dealing with complex topological structures and dynamic fault propagation, and cannot meet the needs of efficient assessment of modern smart distribution networks.

[0004] In recent years, risk assessment methods based on deep learning have gradually attracted attention, especially the use of neural networks to capture the complex dependencies between the operating status of the power grid and fault characteristics. For example, convolutional neural networks can extract the spatial characteristics of the power grid, but their ability to capture temporal dynamic characteristics is limited. Long short-term memory networks can process time series data well, but cannot directly reflect the topological structure of the power grid. For this reason, graph convolutional networks, as a neural network that can process graph structured data, have been used to extract the spatial topological characteristics of the power grid, and combined with LSTM to capture temporal dynamic changes, forming a risk assessment method that has the ability to capture both spatial and temporal characteristics. By integrating these two networks, the global impact of information attacks on CPDS can be more accurately reflected. Summary of the invention

[0005] In view of the problems existing in the existing distribution network information system risk assessment method and system based on graph convolutional network, the present invention is proposed.

[0006] Therefore, the problem to be solved by the present invention is that the convolutional neural network can extract the spatial characteristics of the power grid, but its ability to capture temporal dynamic characteristics is limited, while the long short-term memory network can process time series data well, but cannot directly reflect the topological structure of the power grid. The present invention solves this problem by adopting a distribution network information system risk assessment method based on a graph convolutional network.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In the first aspect, an embodiment of the present invention provides a distribution network information system risk assessment method based on a graph convolutional network, which includes generating the operation data of the distribution network information physical system under information attack through simulation modeling, using GCN to extract the complex topological structure and electrical characteristics of the system, combining LSTM to capture the time dynamic characteristics of the fault propagation process, and accurately assessing the risk level of node voltage exceeding the limit;

[0009] Design a simulation modeling method for controllable resource communication interruption under information attack, generate power grid fault samples through Monte Carlo simulation, and define distribution network risk assessment indicators based on simulation results, taking node voltage over-limit as the core evaluation criterion to comprehensively reflect the risk distribution under different attack scenarios;

[0010] The input data of the GCN-LSTM model is normalized to a uniform length, and the normalization processing technology is combined to enhance the generalization ability of the model, learn the electrical characteristics of the system under non-fault conditions, improve the evaluation accuracy, and complete the risk assessment of the distribution network information system.

[0011] As a preferred solution of the distribution network information system risk assessment method based on graph convolutional network described in the present invention, wherein: the generation of distribution network information physical system by simulation modeling includes attacking the control command downlink communication link, which is completed in the following manner;

[0012] In the CDLA attack, the state of the distribution network communication link is modified, as described in detail:

[0013]

[0014] In the formula, l i-C is the control instruction of the master station to the controllable resources at node i before the attack. is the control instruction of the master station to the controllable resources at node i before the attack;

[0015] CDLA attacks can cause controllable resources such as wind power and photovoltaic power connected to the node to go offline, causing the node voltage to exceed the limit.

[0016] As a preferred solution of the distribution network information system risk assessment method based on graph convolutional network described in the present invention, wherein: the generation of power grid fault samples by Monte Carlo simulation includes simulating different types of faults occurring at different locations and at different time points, and completing the fault analysis caused by information attack through the following steps:

[0017] Initialize system parameters, set generator capacity, generator output, load and power system topology before CDLA attack, randomly generate initial faults at different locations and time points through MCS method, and set the initial time T = 0;

[0018] Power flow calculation and data recording, perform power flow calculation, update time T = T + 1, record the system data of fault N at time T, and determine whether the node voltage exceeds the limit. To simulate the fault repair process, the power flow calculation interval is 15 minutes, and the upper limit of the simulation time is set to T = 8;

[0019] Fault status judgment: if the voltage exceeds the limit, end the simulation of the current fault N and enter fault N+1. If T<8 ​​and the voltage does not exceed the limit, return to the power flow calculation step and continue the simulation.

[0020] As a preferred solution of the distribution network information system risk assessment method based on graph convolutional network described in the present invention, wherein: the definition of distribution network risk assessment index includes defining node voltage over-limit, which is expressed by the following formula:

[0021]

[0022] In the formula, R i is the risk assessment index of node i, U i is the voltage of node i;

[0023] The method of enhancing the generalization ability of the model by combining the normalization processing technology includes adopting data enhancement and normalization technology to improve the generalization ability of the model and determine the uniformity of the input data. The specific steps are as follows:

[0024] The data enhancement includes inserting a non-fault feature graph with a time length of 10-T into the front end of the sequence when the length of the fault sequence is T, so that the time length of all fault graph sequences is standardized to 10;

[0025] The normalization includes standardizing all node data to a data form with a mean of 0 and a standard deviation of 1, thereby enhancing the adaptability of the model to different data distributions.

[0026] As a preferred solution of the distribution network information system risk assessment method based on graph convolutional network described in the present invention, wherein: the electrical characteristics of the learning system under non-fault conditions include building a risk assessment framework by combining graph convolutional network and long short-term memory network, and completing data-driven risk prediction by building a GCN-LSTM model. The model is composed of a graph convolutional network GCN and a long short-term memory network LSTM. The failure graph sequence generated by simulation is used as input, and the topological structure and electrical characteristics of the power system are extracted using GCN to capture the association relationship between nodes. The fault graph sequence containing the topological structure and electrical characteristics is input into the LSTM layer, and the dynamic changes in the fault propagation process are captured through information transmission between LSTM units. The risk assessment results are output, and the topological feature extraction model of the risk assessment framework is completed in combination with the spatiotemporal characteristics of distribution network faults.

[0027] As a preferred solution of the distribution network information system risk assessment method based on graph convolutional network described in the present invention, the topological feature extraction model includes building a topological feature extraction model of the power system using graph convolutional network, and obtaining the topological structure of the system by learning the complex dependency relationship between electrical features. The output of the GCN layer is expressed as:

[0028]

[0029] Where X and Y are the input and output matrices of the GCN layer, respectively. X represents the characteristic matrix of the power system network, each row corresponds to the electrical characteristics of the bus, and A is the adjacency matrix. is the degree matrix, defined as:

[0030]

[0031] Among them, W is the trainable parameter matrix of the GCN layer, I is the identity matrix, and f σ (·) is the activation function. The proposed method uses the active power P of the node i , reactive power Q i , voltage amplitude U i , voltage phase angle θ i , photovoltaic generator power P vi and wind turbine power P wi As a characteristic of the power system status;

[0032] The topology of the acquisition system includes capturing the spatial correlation of the power grid, using a two-layer GCN model, whose output is expressed as:

[0033]

[0034] Among them, X GCN and Y GCNThey represent the input and output of the 2-layer GCN model respectively, and W1 and W2 represent the trainable parameter matrices of the first and second layers of GCN respectively.

[0035] As a preferred solution of the distribution network information system risk assessment method based on graph convolutional network described in the present invention, the specific formula of the long short-term memory network LSTM is:

[0036] f t =σ(W f ·[h t-1 ,x t ]+b f )

[0037] i t =σ(W i ·[h t-1 ,x t ]+b i )

[0038]

[0039] O t =σ(W o ·[h t-1 ,x t ]+b o )

[0040] h t =O t ⊙tanh(C t )

[0041] Among them, f t is the output of the forget gate, σ is the sigmoid activation function, W f is the weight matrix of the forget gate, b f , b i , b c and b o is the bias vector, h t-1 is the hidden state of the previous time step, x t is the input of the current time step, i t is the output of the input gate, W i is the weight matrix of the input gate, is the state of the candidate memory cell, W c is the weight matrix of the candidate memory unit, C t is the cell state at the current time step, C t-1 is the cell state at the previous time step, ⊙ represents the element-wise multiplication of the Hadamard product, O t is the output of the output gate, W o is the weight matrix of the output gate, h tis the hidden state at the current time step.

[0042] In the second aspect, an embodiment of the present invention provides a distribution network information system risk assessment system based on a graph convolutional network, which includes: an extraction module, which generates operating data of a distribution network information-physical system under information attacks through simulation modeling, uses GCN to extract the complex topological structure and electrical characteristics of the system, and combines LSTM to capture the temporal dynamic characteristics of the fault propagation process, and accurately assesses the risk level of node voltage exceeding the limit; a design module, which designs a simulation modeling method for controllable resource communication interruption under information attacks, generates power grid fault samples through Monte Carlo simulation, and defines distribution network risk assessment indicators based on simulation results, taking node voltage exceeding the limit as the core evaluation criterion, and comprehensively reflecting the risk distribution under different attack scenarios; an evaluation module, which normalizes the input data of the GCN-LSTM model to a uniform length, combines normalization processing technology to enhance the generalization ability of the model, learns the electrical characteristics of the system under non-fault conditions, improves the evaluation accuracy, and completes the risk assessment of the distribution network information system.

[0043] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the processor executes the computer program, it implements any step of the above-mentioned distribution network information system risk assessment method based on graph convolutional network.

[0044] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the above-mentioned distribution network information system risk assessment method based on graph convolutional network is implemented.

[0045] The beneficial effects of the present invention are as follows: by introducing a combined model based on graph convolutional network GCN and long short-term memory network LSTM, the present invention can effectively capture the spatiotemporal characteristics in the information-physical system of the distribution network, accurately evaluate the risk level of node voltage exceeding the limit, and significantly improve the accuracy and real-time performance of risk assessment. The present invention can not only identify high-risk nodes, but also provide reliable support for the safe operation of the power grid, thereby improving the safety and resilience of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0047] Figure 1A specific flow chart of a distribution network information system risk assessment method and system based on a graph convolutional network provided for one embodiment of the present invention.

[0048] Figure 2 A GCN-LSTM model flow chart of a distribution network information system risk assessment method and system based on a graph convolutional network provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0050] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0051] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0052] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0053] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0054] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0055] Example 1

[0056] Reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a distribution network information system risk assessment method based on a graph convolutional network, comprising:

[0057] S1: Generate the operation data of the distribution network cyber-physical system under information attack through simulation modeling, use GCN to extract the complex topology and electrical characteristics of the system, combine LSTM to capture the temporal dynamic characteristics of the fault propagation process, and accurately evaluate the risk level of node voltage exceeding the limit.

[0058] Among them, the generation of the distribution network cyber-physical system through simulation modeling includes attacks on the control command downlink communication link, which is completed in the following ways;

[0059] In the CDLA attack, the state of the distribution network communication link is modified, as described in detail:

[0060]

[0061] In the formula, l i-C is the control instruction of the master station to the controllable resources at node i before the attack. is the control instruction of the master station to the controllable resources at node i before the attack;

[0062] CDLA attacks can cause controllable resources such as wind power and photovoltaic power connected to the node to go offline, causing the node voltage to exceed the limit.

[0063] S2: Design a simulation modeling method for controllable resource communication interruption under information attack, generate power grid fault samples through Monte Carlo simulation, and define distribution network risk assessment indicators based on simulation results, taking node voltage over-limit as the core evaluation criterion to comprehensively reflect the risk distribution under different attack scenarios.

[0064] Among them, the generation of power grid fault samples through Monte Carlo simulation includes simulating different types of faults occurring at different locations and different time points, and completing the fault analysis caused by information attacks through the following steps:

[0065] Initialize system parameters, set generator capacity, generator output, load and power system topology before CDLA attack, randomly generate initial faults at different locations and time points through MCS method, and set the initial time T = 0;

[0066] Power flow calculation and data recording, perform power flow calculation, update time T = T + 1, record the system data of fault N at time T, and determine whether the node voltage exceeds the limit. To simulate the fault repair process, the power flow calculation interval is 15 minutes, and the upper limit of the simulation time is set to T = 8;

[0067] Fault status judgment: if the voltage exceeds the limit, end the simulation of the current fault N and enter fault N+1. If T<8 ​​and the voltage does not exceed the limit, return to the power flow calculation step and continue the simulation.

[0068] S2.1: Defining the risk assessment index of the distribution network includes defining the node voltage limit, which is expressed by the following formula:

[0069]

[0070] In the formula, R i is the risk assessment index of node i, U i is the voltage of node i;

[0071] Combining normalization processing technology to enhance the generalization ability of the model includes using data enhancement and normalization technology to improve the generalization ability of the model and determine the uniformity of the input data. The specific steps are as follows:

[0072] Data augmentation includes inserting a non-fault feature graph with a time length of 10-T into the front end of the sequence when the length of the fault sequence is T, so that the time length of all fault graph sequences is standardized to 10;

[0073] Normalization involves standardizing all node data into a data form with a mean of 0 and a standard deviation of 1, thereby enhancing the model's adaptability to different data distributions.

[0074] S3: Normalize the input data of the GCN-LSTM model to a uniform length, combine the normalization processing technology to enhance the generalization ability of the model, learn the electrical characteristics of the system under non-fault conditions, improve the assessment accuracy, and complete the risk assessment of the distribution network information system.

[0075] Among them, learning the electrical characteristics of the system under non-fault conditions includes building a risk assessment framework by combining graph convolutional networks and long short-term memory networks, and completing data-driven risk prediction by constructing a GCN-LSTM model. The model is composed of a graph convolutional network GCN and a long short-term memory network LSTM. The failure graph sequence generated by simulation is used as input, and GCN is used to extract the topological structure and electrical characteristics of the power system, thereby capturing the association between nodes. The fault graph sequence containing topological structure and electrical characteristics is input into the LSTM layer, and the dynamic changes in the fault propagation process are captured through information transmission between LSTM units. The risk assessment results are output, and the topological feature extraction model of the risk assessment framework is completed in combination with the spatiotemporal characteristics of distribution network faults.

[0076] S3.1: The topological feature extraction model includes building a topological feature extraction model of the power system using a graph convolutional network. The topological structure of the system is obtained by learning the complex dependencies between electrical features. The output of the GCN layer is expressed as:

[0077]

[0078] Where X and Y are the input and output matrices of the GCN layer, respectively. X represents the characteristic matrix of the power system network, each row corresponds to the electrical characteristics of the bus, and A is the adjacency matrix. is the degree matrix, defined as:

[0079]

[0080] Among them, W is the trainable parameter matrix of the GCN layer, I is the identity matrix, and f σ (·) is the activation function. The proposed method uses the active power P of the node i , reactive power Q i , voltage amplitude U i , voltage phase angle θ i , photovoltaic generator power P vi and wind turbine power P wi As a characteristic of the power system status;

[0081] Obtaining the topological structure of the system includes capturing the spatial correlation of the power grid. A two-layer GCN model is used, and its output is expressed as:

[0082]

[0083] Among them, X GCN and Y GCN They represent the input and output of the 2-layer GCN model respectively, and W1 and W2 represent the trainable parameter matrices of the first and second layers of GCN respectively.

[0084] S3.2: The specific formula of the long short-term memory network LSTM is:

[0085] f t =σ(W f ·[h t-1 ,x t ]+b f )

[0086] i t =σ(W i ·[h t-1 ,x t ]+b i )

[0087]

[0088] O t =σ(W o ·[h t-1 ,x t ]+b o )

[0089] h t =O t ⊙tanh(C t )

[0090] Among them, f t is the output of the forget gate, σ is the sigmoid activation function, W f is the weight matrix of the forget gate, b f , b i , b c and b o is the bias vector, h t-1 is the hidden state of the previous time step, x t is the input of the current time step, i t is the output of the input gate, W i is the weight matrix of the input gate, is the state of the candidate memory cell, W c is the weight matrix of the candidate memory unit, C t is the cell state at the current time step, C t-1 is the cell state at the previous time step, ⊙ represents the element-wise multiplication of the Hadamard product, O t is the output of the output gate, W o is the weight matrix of the output gate, h t is the hidden state at the current time step.

[0091] In a preferred embodiment, a distribution network information system risk assessment system based on a graph convolutional network includes an extraction module, which generates operating data of the distribution network information-physical system under information attacks through simulation modeling, uses GCN to extract the complex topological structure and electrical characteristics of the system, and combines LSTM to capture the temporal dynamic characteristics of the fault propagation process to accurately assess the risk level of node voltage exceeding the limit; a design module, which designs a simulation modeling method for controllable resource communication interruption under information attacks, generates power grid fault samples through Monte Carlo simulation, and defines distribution network risk assessment indicators based on simulation results, taking node voltage exceeding the limit as the core evaluation criterion, and comprehensively reflecting the risk distribution under different attack scenarios; an evaluation module, which normalizes the input data of the GCN-LSTM model to a uniform length, combines normalization processing technology to enhance the generalization ability of the model, learns the electrical characteristics of the system under non-fault conditions, improves the evaluation accuracy, and completes the risk assessment of the distribution network information system.

[0092] The above-mentioned unit modules may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to the above-mentioned modules.

[0093] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0094] In summary, the present invention, by introducing a combined model based on graph convolutional network GCN and long short-term memory network LSTM, can effectively capture the spatiotemporal characteristics in the information-physical system of the distribution network, accurately assess the risk level of node voltage exceeding the limit, and significantly improve the accuracy and real-time performance of risk assessment. The present invention can not only identify high-risk nodes, but also provide reliable support for the safe operation of the power grid, thereby improving the safety and resilience of the distribution network.

[0095] Example 2

[0096] Reference Figure 1 and Figure 2 , which is the second embodiment of the present invention, provides a distribution network information system risk assessment method based on graph convolutional network. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0097] A simulated distribution network model with 14 nodes was constructed. The model can generate operating data under different information attack scenarios. For attacks on the control command downlink communication link, the state of the distribution network communication link is modified to simulate the changes in the control instructions of the master station to the controllable resources at the node before and after the attack. Specifically, the attack caused the controllable resources such as wind power and photovoltaic power to be disconnected, which in turn caused the node voltage to exceed the limit. Through the Monte Carlo simulation method, we generated 10,000 sets of power grid fault samples. These samples covered different types of faults occurring at different locations and at different time points. In each simulation cycle, the system parameters were initialized, including the generator capacity of 50MW, the output power of 30MW, the load of 40MW, and the topology of the power system.

[0098] The power flow calculation is performed in each 15-minute interval, the time T=T+1 is updated, the system data of fault N at time T is recorded, and it is determined whether the node voltage exceeds the limit. The upper limit of the simulation time is set to 8 hours. For each fault scenario, if the node voltage exceeds the limit at any time point, which is defined as the voltage exceeds the rated value by ±10%, the simulation of the current fault N is immediately terminated and the simulation of the next fault N+1 is started; if the voltage does not exceed the limit within 8 hours, the simulation continues until the upper limit is reached.

[0099] In order to train and test the GCN-LSTM model, all fault samples are divided into 70% training set, 15% validation set and 15% test set. The model input data is normalized to a uniform length, that is, the time length of all fault sequences is standardized to 10 time steps, and the non-fault feature map is inserted into the front of the sequence to fill it to 10 time steps. In addition, we normalized all node data and converted it into a data form with a mean of 0 and a standard deviation of 1 to enhance the generalization ability of the model. Finally, the accuracy of the GCN-LSTM model on the test set reached 92.3%, indicating that the model can effectively predict the risk level of node voltage exceeding the limit in the distribution network under information attack, thereby providing important decision support for the safe and stable operation of the power grid. The experimental data is shown in Table 1 below:

[0100] Table 1 Experimental data table

[0101]

[0102] Table 1 constructs a simulated distribution network model, simulates the operation of the power grid under information attack, generates a large number of fault samples, and performs risk assessment through power flow calculation and voltage over-limit judgment. After the data is standardized, the GCN-LSTM model is used for training and testing, and the effectiveness of the model in predicting the risk of node voltage over-limit is finally verified, providing decision support for the safe and stable operation of the power grid. The comparison between the present invention and the prior art is shown in Table 2 below:

[0103] Table 2 Comparison between the present invention and the prior art

[0104]

[0105] Table 2 shows significant advantages in terms of risk assessment accuracy, data processing automation, topological structure capture, time dynamics consideration, and model generalization ability.

[0106] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A distribution network information system risk assessment method based on graph convolutional network, characterized by: include, Through simulation modeling, the cyber-physical system of the distribution network is generated, and the operating data under information attacks is extracted by GCN. The complex topological structure and electrical characteristics of the system are extracted, and the temporal dynamic characteristics of the fault propagation process are captured by LSTM to accurately assess the risk level of node voltage exceeding the limit. Design a simulation modeling method for controllable resource communication interruption under information attack, generate power grid fault samples through Monte Carlo simulation, and define distribution network risk assessment indicators based on simulation results, taking node voltage over-limit as the core evaluation criterion to comprehensively reflect the risk distribution under different attack scenarios; The input data of the GCN-LSTM model is normalized to a uniform length, and the normalization processing technology is combined to enhance the generalization ability of the model, learn the electrical characteristics of the system under non-fault conditions, improve the evaluation accuracy, and complete the risk assessment of the distribution network information system.

2. The distribution network information system risk assessment method based on graph convolutional network according to claim 1, characterized in that: The generation of the power distribution network cyber-physical system by simulation modeling includes attacking the control command downlink communication link, which is accomplished by the following methods; In the CDLA attack, the state of the distribution network communication link is modified, as described in detail: In the formula, l i-C is the control instruction of the master station to the controllable resources at node i before the attack. is the control instruction of the master station to the controllable resources at node i before the attack; CDLA attacks can cause the wind power and photovoltaic controllable resources connected to the node to go offline, causing the node voltage to exceed the limit.

3. The distribution network information system risk assessment method based on graph convolutional network according to claim 2, characterized in that: The generation of power grid fault samples by Monte Carlo simulation includes simulating different types of faults occurring at different locations and at different time points, and completing the fault analysis caused by information attacks by the following steps: Initialize system parameters, set generator capacity, generator output, load and power system topology before CDLA attack, randomly generate initial faults at different locations and time points through MCS method, and set the initial time T = 0; Power flow calculation and data recording, perform power flow calculation, update time T = T + 1, record the system data of fault N at time T, and determine whether the node voltage exceeds the limit. To simulate the fault repair process, the power flow calculation interval is 15 minutes, and the upper limit of the simulation time is set to T = 8; Fault status judgment: if the voltage exceeds the limit, end the simulation of the current fault N and enter fault N+1. If T<8 ​​and the voltage does not exceed the limit, return to the power flow calculation step and continue the simulation.

4. The distribution network information system risk assessment method based on graph convolutional network as claimed in claim 3, characterized in that: Defining the distribution network risk assessment index includes defining node voltage over-limit, which is expressed by the following formula: In the formula, R i is the risk assessment index of node i, U i is the voltage of node i; The method of enhancing the generalization ability of the model by combining the normalization processing technology includes adopting data enhancement and normalization technology to improve the generalization ability of the model and determine the uniformity of the input data. The specific steps are as follows: The data enhancement includes inserting a non-fault feature graph with a time length of 10-T into the front end of the sequence when the length of the fault sequence is T, so that the time length of all fault graph sequences is standardized to 10; The normalization includes standardizing all node data to a data form with a mean of 0 and a standard deviation of 1, thereby enhancing the adaptability of the model to different data distributions.

5. The distribution network information system risk assessment method based on graph convolutional network according to claim 4, characterized in that: The electrical characteristics of the learning system under non-fault conditions include building a risk assessment framework by combining a graph convolutional network with a long short-term memory network, completing data-driven risk prediction by building a GCN-LSTM model, the model is composed of a graph convolutional network GCN and a long short-term memory network LSTM, taking a failure graph sequence generated by simulation as input, using GCN to extract the topological structure and electrical characteristics of the power system, thereby capturing the association between nodes, inputting the fault graph sequence containing the topological structure and electrical characteristics into the LSTM layer, capturing dynamic changes in the fault propagation process through information transmission between LSTM units, outputting risk assessment results, and combining the spatiotemporal characteristics of distribution network faults to complete the topological feature extraction model of the risk assessment framework.

6. The distribution network information system risk assessment method based on graph convolutional network according to claim 5, characterized in that: The topological feature extraction model includes building a topological feature extraction model of the power system using a graph convolutional network, and obtaining the topological structure of the system by learning the complex dependencies between electrical features. The output of the GCN layer is expressed as: Where X and Y are the input and output matrices of the GCN layer, respectively. X represents the characteristic matrix of the power system network, each row corresponds to the electrical characteristics of the bus, and A is the adjacency matrix. is the degree matrix, defined as: Among them, W is the trainable parameter matrix of the GCN layer, I is the identity matrix, and f σ (·) is the activation function. The proposed method uses the active power P of the node i , reactive power Q i , voltage amplitude U i , voltage phase angle θ i , photovoltaic generator power P vi and wind turbine power P wi As a characteristic of the power system status; The topology of the acquisition system includes capturing the spatial correlation of the power grid, using a two-layer GCN model, whose output is expressed as: Among them, X GCN and Y GCN They represent the input and output of the 2-layer GCN model respectively, and W1 and W2 represent the trainable parameter matrices of the first and second layers of GCN respectively.

7. The distribution network information system risk assessment method based on graph convolutional network according to claim 6, characterized in that: The specific formula of the long short-term memory network LSTM is: f t =σ(W f ·[h t-1 ,x t ]+b f ) i t =σ(W i ·[h t-1 ,x t ]+b i ) The t =σ(W o ·[h t-1 ,x t ]+b o ) h t =O t ⊙tanh(C t ) Among them, f t is the output of the forget gate, σ is the sigmoid activation function, W f is the weight matrix of the forget gate, b f , b i , b c and b o is the bias vector, h t-1 is the hidden state of the previous time step, x t is the input of the current time step, i t is the output of the input gate, W i is the weight matrix of the input gate, is the state of the candidate memory cell, W c is the weight matrix of the candidate memory unit, C t is the cell state at the current time step, C t-1 is the cell state at the previous time step, ⊙ represents the element-wise multiplication of the Hadamard product, O t is the output of the output gate, W o is the weight matrix of the output gate, h t is the hidden state at the current time step.

8. A distribution network information system risk assessment system based on a graph convolutional network, based on the distribution network information system risk assessment method based on a graph convolutional network according to any one of claims 1 to 7, characterized in that: include, The extraction module generates the operation data of the distribution network cyber-physical system under information attacks through simulation modeling, uses GCN to extract the complex topological structure and electrical characteristics of the system, and combines LSTM to capture the temporal dynamic characteristics of the fault propagation process to accurately assess the risk level of node voltage exceeding the limit; The design module designs a simulation modeling method for controllable resource communication interruption under information attack, generates power grid fault samples through Monte Carlo simulation, and defines distribution network risk assessment indicators based on simulation results, taking node voltage over-limit as the core evaluation criterion to comprehensively reflect the risk distribution under different attack scenarios; The evaluation module normalizes the input data of the GCN-LSTM model to a uniform length, combines normalization processing technology to enhance the generalization ability of the model, learns the electrical characteristics of the system under non-fault conditions, improves the evaluation accuracy, and completes the risk assessment of the distribution network information system.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the distribution network information system risk assessment method based on graph convolutional network described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the distribution network information system risk assessment method based on graph convolutional network described in any one of claims 1 to 7 are implemented.

Images