Power system attack and defense simulation method and equipment based on abstract topological structure

Through the power system attack and defense simulation method based on abstract topology, combined with particle swarm algorithm and hierarchical analysis method, multiple network attack scenarios are simulated, and the problem of separation of power system and network layer in the existing technology is solved, and the complex network of power system is realized Offensive and defensive simulation improves the safety and stability of the power system.

CN119026317AInactive Publication Date: 2024-11-26WUHAN UNIV

Patent Information

Application Number
CN202410945036.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-15
Publication Date
2024-11-26
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing network shooting range is difficult to simulate multiple network attack methods, cannot meet the complex network security needs of the power industry, and the existing technology cannot effectively connect the power layer with the network layer, limiting the network attack and defense simulation capabilities of the power system.

Method used

Adopt the power system attack and defense simulation method based on abstract topology, and by establishing an information network model and a power information fusion system model, combining particle swarm algorithm and hierarchical analysis method, multiple network attack scenarios are simulated, and the vulnerability and security status of the power system are evaluated. Achieve the combination of power system and network layer.

Benefits of technology

It provides a risk-free environment that allows security teams to conduct practical offensive and defense drills, improve emergency response capabilities and practical experience, identify potential vulnerabilities and risks, enhance the security and stability of the power system, and improve network security protection capabilities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119026317A_ABST
    Figure CN119026317A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of power system security, and particularly relates to a power system attack and defense simulation method and device based on an abstract topological structure, and the method comprises the steps: building an information network model and a power information fusion system model; building a network simulation platform of the power system; the method comprises the following steps of: comprehensively considering a topological structure and an operation state of a system, selecting a vulnerability index, defining a coincidence loss # imgabs0 # based on load importance, selecting a vulnerability index of a power system, and establishing a power system state evaluation system by utilizing an analytic hierarchy process based on a safety and stability theory of the power system; and reconstructing and modeling the power distribution network. The method maintains the safety and stability of the power system, identifies the potential safety risk of the power system through testing and drilling, and formulates a corresponding risk coping strategy. The safety condition of the electric power system can be better known, problems can be timely found and improved, and the safety of the electric power system is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of power system security, and in particular relates to a power system attack and defense simulation method and equipment based on an abstract topology structure. Background Art

[0002] Whether it is the government, enterprises, or public institutions, they have invested a lot of financial, material and human resources in the construction of a practical environment for cyber security attack and defense. The research and development of cyber ranges can be roughly divided into the following three stages: Phase 1: The physical target cyber range period in the early 21st century. The range format at this stage was relatively simple and direct. Each party would usually use the enemy's network equipment as a target to build a physical target cyber range. Through this range, the latest attack tools of the party were tested to evaluate their "power", that is, whether they could bypass the opponent's security protection system. In addition, while discovering the enemy's vulnerabilities, the party would also check whether there were similar vulnerabilities on the party and repair them in a timely manner. Typical representatives include the early international hacker attack and defense wars and Trojan test systems.

[0003] The second stage: the small virtualized network range period. This stage began around 2005. The ranges during this period began to use cloud computing, software-defined networks, and virtualization technologies to simulate real network environments for network attack and defense. However, the network ranges during this stage were small in scale, and although a certain degree of network attack and defense drills could be conducted, there were still many limitations.

[0004] The third stage: the period of large-scale virtual and real combined cyber ranges and industry-specific cyber ranges. This stage began around 2014. During this stage, the scale and complexity of cyber ranges increased significantly, and they were able to support large-scale complex cyber attack and defense confrontation exercises. At the same time, with the emergence of new cyber attack methods, network security issues emerged in an endless stream, and the importance of network security has therefore received great attention. During this stage, there were also special cyber ranges for specific industries to meet the network security needs of different industries. In recent years, some of the research results of the network range mainly include the following: Stephen G Batse and others developed a security framework that integrates existing network security systems to provide real-time situational awareness of large-scale networks, but this method has great limitations in deployment; Jason Shifflett used ontology to analyze and compare concepts related to network security situational awareness and built a modular framework structure, but this method can only detect limited, known attacks and has not yet achieved network security threat recognition.

[0005] Dong Hui and Ma Jian's "Construction of a Network Attack and Defense Experimental Platform Based on Virtual Honeynet". This study pointed out some shortcomings of the traditional passive defense network range, focused on the honeypot technology, and finally used virtual machines to build a network range based on honeypot technology. This range can simulate network attack behaviors and defense measures, and is particularly suitable for network security teaching experiments, but it has limitations in its application in the power industry.

[0006] Kong Hongshan, Tang Jun and Zhang Mingqing's "Design and Implementation of a SITL-Based Network Attack and Defense Simulation Platform". This study uses SITL (Software in the Loop) technology when building a network range to address the lack of authenticity in network attack and defense simulations. It overcomes the difficulties in modeling network attack models and designs and implements the simulation process of the simulation platform. However, the network range can only simulate a few limited network attack methods such as ARP attacks and SYN Flood, and the response to a variety of common network attack methods needs to be further improved.

[0007] "Smart Grid Reliability Assessment Method Considering Communication Architecture Failure Risk" by Zhu Wentao, Han Mingyu, Jovica Milanovic, and Peter Crossley proposed an analysis method based on complex network theory. It introduces the delay time of data packets and the reliability of communication equipment to evaluate the reliability of communication paths and the risk of grid failure due to network failures. Finally, two innovative indicators, system reliability improvement value and deterioration value, are used to conduct sensitivity analysis on various components of the power communication network.

[0008] Cao Ge, Gu Wei, Li Peixin, Sheng Wanxing, and Liu Keyan's "Operational Risk Assessment of Active Distribution Network Considering Network Emergency Events" studied a quantitative assessment method for operational risk based on power parameter-communication network joint analysis, which is mainly used to analyze the information flow and network emergency events of the network system. However, existing academic research rarely links to actual industry security needs. These studies often focus on the quantitative modeling of risk indicators and ignore the assessment of real abnormal events on the network side.

[0009] Most of the research on network attack and defense ranges uses virtualization technology to simulate the actual network environment. However, this type of experimental environment can only conduct network attack and defense under limited conditions, and its versatility is obviously insufficient. For example, they can usually only deal with specific types of network attacks. Universities and research institutions often conduct research from the perspective of theoretical models, while network security companies are more focused on solving actual security needs. In general, although virtualization technology has made some progress in the research of network ranges, the current experimental environment still needs to be improved in terms of wide applicability and comprehensiveness to better cope with various complex network attack scenarios. At present, the commonly used electromechanical transient models abroad include PSD-BPA, PSASP, PSS / E, NETOMAC, Digsilent, etc.; and the electromagnetic transient simulation software mainly includes EMTP / ATP, EMTDC / PSCAD, NETOMAC, RTDS, RTLAB, etc. CyberBattleSim developed by Microsoft can simulate the confrontation between attackers and defenders at the network layer, and the power industrial control network attack and defense range can simulate power layer links such as power generation, transmission, substation and distribution, but there is currently no software that can link the power layer with the network layer, which is also the focus of our research and development. Summary of the invention

[0010] The present invention provides an attack and defense simulation method and device for an electric power system based on an abstract topology structure, so as to solve the defect that the network target range in the prior art can only simulate limited network attack methods and cannot solve the actual security needs, realize the combination of network attack and defense with the power grid information system, and carry out actual network attack and defense in the electric power industry.

[0011] The present invention provides a power system attack and defense simulation method based on an abstract topology structure, comprising the following steps: Based on the steady-state power flow equation of the power grid, establish the information network model and the power information fusion system model; Build a network simulation platform for the power system, select the standard IEEE33 node system for power flow calculation and topology analysis, and verify the effectiveness of the model through case analysis; Considering the topology structure and the system operation status, select the information node loss ratio , the relative transmission efficiency of information networks As a vulnerability indicator of the information layer, a compliance loss based on load importance is defined. , select power system vulnerability indicators, and establish a power system status assessment system using the analytic hierarchy process based on power system security and stability theory; Distribution network reconstruction modeling. Establish the IEEE33 node distribution network model, establish the distribution network reconstruction model and ordered ring network matrix coding based on the particle swarm algorithm, and implement the distribution network reconstruction code programming in MATLAB.

[0012] According to a power system attack and defense simulation method based on an abstract topology structure provided by the present invention, the establishment of an information network model and a power information fusion system model based on a power grid steady-state power flow equation includes the following steps: 1.1) Establishing an information network model includes: The power information network is abstracted into a point set and edge sets The composition of the picture , the information transmission equipment is abstracted as the nodes of the graph, and the optical fiber transmission route is abstracted as the edges of the graph; based on the power information network, abstract modeling is performed to obtain a power information network model composed of different nodes and edges; 1.2) Modeling of power information fusion system: The power information fusion system is abstracted as a graph composed of point sets and edge sets, including different nodes and branches. The nodes are the nodes in the connection layer, backbone layer and core layer, and the branches are the branches in the power network and information network.

[0013] According to a power system attack and defense simulation method based on an abstract topology structure provided by the present invention, the network simulation platform of the power system is built, a standard IEEE33 node system is selected for power flow calculation and topology analysis, and the validity of the model is verified by example analysis, including the following steps: 2.1) Build a power system network simulation platform; Through platform virtualization, a logically exclusive computing environment is provided for nodes in the power system simulation network, and through network virtualization, logically independent network devices are provided for the construction of network links; 2.2) Perform power flow calculation and topology analysis; The constructed model is calculated using the MATLAB power flow calculation program, and the optimal network reconstruction method is obtained using the genetic algorithm.

[0014] According to a power system attack and defense simulation method based on an abstract topology structure provided by the present invention, the selected information node loss ratio , the relative transmission efficiency of information networks As a vulnerability indicator of the information layer, a compliance loss based on load importance is defined. , select power system vulnerability indicators, based on the power system security and stability theory, use the hierarchical analysis method to establish a power system status assessment system including: 3.1) Information node loss ratio for,

[0015] In the formula Represents the number of initial nodes of the information network; Represents the number of remaining nodes of the information node after the information system is attacked; 3.2) Select the relative transmission efficiency of the information network include: Information network transmission efficiency for,

[0016] In the formula Represents the total number of nodes in the information network; Represents data through the shortest path from the source node Send to Node The delay caused by the process; Define the relative transmission efficiency of information networks for,

[0017] In the formula Represents the network transmission efficiency when the information network is not attacked; The value range is ,when When , it means that after the information network is attacked, the network transmission efficiency does not decrease; when When , it means that after the information network is attacked, the network transmission efficiency drops to zero; 3.3) Define load loss based on load importance include: Define significant load reduction as load loss ,

[0018] In the formula Represents load node Importance weight of load; Represents the active power of the power node before load reduction; Represents the active power of the power node after load reduction; 3.4) Selected power system vulnerability indicators include: 3.4.1) Consequence hazard indicators; Set Attack Cause a change in a certain node or line while ignoring the intermediate process, attack The index of the number of unstable nodes caused by

[0019] In the formula For attack The number of unstable nodes caused, is the number of stable nodes in the power grid topology; 3.4.2) The basic recovery time index, i.e. the recovery time index of the unstable node, is: attack The recovery time index of the unstable node caused by , which reflects the power system's vulnerability to attacks The strength of the recovery capability is inversely proportional to the recovery time of the unstable node; 3.4.3) Key indicators of operating status include: Node connection status indicators; Attack The next node connection status indicator is,

[0020] In the formula is the number of connected nodes when there is no attack, For attack The number of downstream nodes; Branch connection status indicator; attack The lower branch connection status indicator is,

[0021] In the formula is the number of all lines when there is no attack, For attack The number of valid lines. Load fluctuation status indicator; attack The load fluctuation state index is:

[0022] In the formula is the base value of load power when there is no attack, For attack Real-time value of load power; 3.5) Establishment of power system status assessment system including: The analytic hierarchy process is used to quantitatively evaluate the power system status, calculate the weight and score of each level indicator, and obtain a comprehensive score based on the weight and score. The comprehensive score is used to reflect the power system status. The specific steps are as follows: 3.5.1) Construct a weight matrix; the set of indicators at a certain level is , is the weight vector for obtaining the index , construct the judgment matrix; use Indicators and For upper level indicators The importance of ; The value of is expressed by the scale of the hierarchical analysis method; the importance of each element at the same level with respect to a criterion at the previous level is compared pairwise to obtain the judgment matrix , perform consistency check on the constructed judgment matrix. If the consistency ratio , then the consistency of the judgment matrix is ​​acceptable, otherwise the judgment matrix needs to be modified; 3.5.2) Weight calculation: The geometric mean method is used to calculate the weight. The weight vector of a certain layer is ,in, , thereby calculating the importance ranking weight of the lowest level indicators; 3.5.3) Calculation of index scores; the index scores of power systems are recorded as ,in , is the raw score of each indicator, , is the full score; 3.5.4) Calculation of comprehensive score; the comprehensive score is the indicator score multiplied by the weight, recorded as ; The higher the comprehensive score, the better the power system status; 3.6) Weak links in the power system include: 3.6.1) Attack scenarios on the power generation side; 3.6.2) Transmission-side attack scenario; 3.6.3) Attack scenario on the power distribution side; 3.6.4) Dispatching automation system.

[0023] According to a power system attack and defense simulation method based on an abstract topology structure provided by the present invention, the distribution network network reconstruction modeling includes: firstly establishing an IEEE33 node distribution network model, establishing a distribution network network reconstruction model and an ordered ring network matrix encoding according to a particle swarm algorithm, and implementing distribution network network reconstruction code programming in matlab; the specific steps are as follows: 4.1) Distribution network reconstruction modeling includes: The ordered ring network coding method is adopted, and the reconstruction of the distribution network is solved based on the particle swarm algorithm; the particle swarm algorithm is used to solve the IEEE33 distribution network formula as follows:

[0024] In the formula, i represents the number of the branch; T is the set of closed branches in the distribution network; For the i The resistance of the branch; For the i The effective value of the voltage at the end node of the branch; and Respectively i Active and reactive power flowing through the end of each branch; The constraints are as follows: (1) Node voltage constraints:

[0025] (2) Branch capacity constraints:

[0026] (3) Connected radial structure constraints:

[0027] Where: , , are the actual voltage of the end node of the i-th branch and its upper and lower limits respectively; , are the actual power flowing through the i-th branch and the maximum power allowed to flow through it; g is the network topology after the distribution network is reconstructed; G is the set of connected radial topology structures; 4.2) Ordered ring network matrix coding includes: 4.2.1) Label each branch in the system, 0 represents the switch on the branch is open, 1 represents the switch on the branch is closed; 4.2.2) Use graph theory to determine the basic loops. The number of loops is the same as the number of tie switches. 4.2.3) List the ring network matrix based on the basic rings. The number of rows in the ring network matrix is ​​the number of basic rings, the number of columns is the number of branches of the ring with the largest number of branches, and the other rings are filled with 0.

[0028] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements any of the above-mentioned power system attack and defense simulation methods based on an abstract topology structure.

[0029] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the power system attack and defense simulation method based on an abstract topology structure as described in any one of the above is implemented.

[0030] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the above-mentioned power system attack and defense simulation methods based on an abstract topology structure.

[0031] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a risk-free environment, allowing the security team to conduct practical attack and defense drills without affecting the actual operation of the power system. By simulating real attack scenarios, security personnel can better understand the attacker's attack methods and attack paths, so as to better formulate defense strategies. This practical drill can improve the emergency response capabilities and practical experience of security personnel, and prepare for real network attacks. Through the drill, a large amount of data about the power system in the face of network attacks can be collected, which is crucial for conducting power system security assessments. Security assessments can be conducted based on these data to determine the system's safety performance level and identify security links that need to be improved or reinforced. In addition, this assessment can also provide a basis for formulating or adjusting security strategies.

[0032] The present invention can simulate various network attack scenarios, help discover vulnerabilities in the power system, and maintain the security and stability of the power system. Through testing and drills, potential security risks of the power system can be identified and corresponding risk response strategies can be formulated. This helps to better understand the security status of the power system, discover problems in a timely manner and make improvements, and improve the security of the power system. The construction of a power system network security range can help power companies improve their network security protection capabilities and improve the security and stability of the power system. Through testing and drills in the range, companies can verify the effectiveness of security measures, repair vulnerabilities in a timely manner, and improve the system's ability to resist attacks.

[0033] The present invention comprehensively considers the topological structure and the operating status of the system, selects the information node loss ratio and the relative transmission efficiency of the information network as the vulnerability indicators of the information layer, and establishes a power system status assessment system using the hierarchical analysis method based on the power system safety and stability theory, which is convenient for intuitively displaying the effectiveness of network attacks and defense measures.

[0034] The present invention adds an attack defense and optimization model, especially the use of a particle swarm algorithm and a defense strategy based on network reconstruction, which can effectively improve the resilience and security of the power system and reduce the impact of network attacks when facing complex network attacks. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the present invention or 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 some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 It is the IEEE33 node model of the power system provided by the embodiment of the present invention; Figure 2It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] This embodiment provides a power system attack and defense simulation method based on an abstract topology structure, which can simulate a variety of network attack scenarios, help discover vulnerabilities in the power system, and maintain the security and stability of the power system. Through testing and drills, potential security risks in the power system can be identified and corresponding risk response strategies can be formulated. This helps to better understand the security status of the power system, discover problems in a timely manner and make improvements, and improve the security of the power system.

[0039] The method of this embodiment has the simulation capability to cover all aspects of power generation, transmission, transformation, distribution, and consumption, involving multiple mainstream attack modes such as FDI, DoS, and short-circuit attacks. It subverts the passive defense research platform model of traditional network security simulation platforms and most honeypot-type platforms, proactively "predicts" potential risks of the power system, and builds a power system network security simulation method that supports business function requirements such as attack and defense confrontation, inspection and testing, emergency drills, and scientific and technological research and development.

[0040] The technical solution adopted in this embodiment is to perform power flow calculation based on IEEE33 nodes, establish a network attack simulation security risk assessment system, and perform a method for power system attack and defense drill simulation, including the following steps: Step 1: Based on the steady-state power flow equation of the power grid, considering the communication transmission process and information decision-making process, the power grid, communication network and information network were hierarchically modeled, and the theoretical modeling of information network modeling and power information fusion system modeling was completed.

[0041] Step 2: Build a network simulation platform suitable for power systems. Based on the standard IEEE33 node system, perform power flow calculation and topology analysis methods, and verify the effectiveness of the provided model through case analysis.

[0042] Step 3: Consider the topology and system operation status and select the information node loss ratio , the relative transmission efficiency of information networks As a vulnerability indicator of the information layer, a compliance loss based on load importance is defined. Based on the power system vulnerability index, the power system status assessment system was established using the analytic hierarchy process based on the power system security and stability theory, which is convenient for intuitively displaying the effectiveness of network attacks and defense measures. Finally, the weak links of the power system were sorted out and summarized.

[0043] Step 4: Use the PyQt5 toolkit in Python to design the software UI interface, complete the code of each model and specific algorithm in Python, and use typical scenarios to test the performance of the software. After the test passes, package the software.

[0044] Step 5: Establish the distribution network reconstruction model. First, establish the IEEE33-node distribution network model. According to the idea of ​​particle swarm algorithm, establish the distribution network reconstruction model and ordered ring network matrix coding. Finally, implement the distribution network reconstruction code programming in MATLAB.

[0045] The implementation steps of step 1 include the following steps: Step 1.1: Information network modeling scheme: In order to meet the information transmission needs of the power system, the power information network is built using technologies such as SDH and optical fiber. These networks are based on different power grid architectures and change with the changes in the power grid architecture. Despite this, most power information networks have some common characteristics: they are composed of a connection layer, a backbone layer, and a core layer. Among these layers, the connection layer is responsible for information collection and the issuance of dispatching commands, and is mainly composed of substations and power plants; the backbone layer is responsible for collecting and transmitting information provided by the connection layer, and is mainly composed of hub substations in the information network; the core layer is responsible for arranging the dispatching plan and operation plan of the power grid according to the rules of the power system, and is mainly composed of the dispatching center.

[0046] Similar to power network modeling, power information network can basically be abstracted as a set of points. and edge sets The composition of the picture , information transmission equipment is abstracted as nodes of the graph, such as dispatching equipment, information collection equipment and routing equipment. Optical fiber transmission routes can be abstracted as edges of the graph. Based on the abstract modeling of the power information network, a power information network model composed of different nodes and edges can be obtained, that is: Step 1.2: Modeling of power information fusion system: Based on the modeling scheme of power network and information network, a modeling scheme of power network integrated security system can be proposed to establish a one-to-one correspondence between power network and information network nodes.

[0047] The power network fusion security system is composed of different nodes and branches, where the nodes are mainly nodes in the connection layer, backbone layer and core layer, and the branches are composed of branches in the power network and information network. The fusion system can be abstracted as a graph composed of a point set and an edge set.

[0048] Since the nodes in the power network correspond to the connection layer, they can communicate through branches to form a fusion system. Therefore, the power network can transmit energy through the fusion system, and the information network can improve the safety of energy transmission through information collection, transmission and analysis, and display the relevant data of the power network fusion safety system at different times.

[0049] The implementation of step 2 includes the following steps: Step 2.1, applicable to the construction of power system network simulation platform; The platform virtualization technology and network virtualization technology are used to realize the power system network simulation. Platform virtualization provides a logically exclusive computing environment for nodes in the power system simulation network, and network virtualization provides logically independent network devices for the construction of network links.

[0050] Step 2.2, perform power flow calculation and topology analysis methods; The data of the constructed model is calculated through the MATLAB power flow calculation program, and the optimal network reconstruction method is obtained using genetic algorithms, which is the main defense measure.

[0051] The implementation of step 3 includes the following steps: Step 3.1: Calculate the information node loss ratio : The transmission of information flow depends on the topological structure of the information network, using the information node loss ratio Measures the vulnerability of the system. When the information system is attacked, the proportion of information nodes lost The larger it is, the more vulnerable the system is and the greater the damage to the topology of the information network.

[0052]

[0053] In the formula Represents the number of initial nodes of the information network; Represents the number of remaining nodes of the information node after the information system is attacked.

[0054] Step 3.2: Find the relative transmission efficiency of the information network : The information system is responsible for data exchange and processing. When the information node is overloaded or failed, it will cause congestion in the information layer, which will further cause information transmission delay. In this case, the issuance of control instructions and the acquisition of basic working status information of the power grid will be affected to varying degrees, thus threatening the stable and reliable operation of the power system.

[0055] The data communication transmission delay is calculated as follows: data transmission in the information network complies with the shortest path principle, and only the delay time of the information node processing data is considered. Each time it passes through an information node, the data delay of the corresponding node is increased. Therefore, the transmission efficiency of the information network : In the formula Represents the total number of nodes in the information network; Represents data through the shortest path from the source node Send to Node Therefore, the transmission efficiency of the information network is :

[0056] In the formula Represents the total number of nodes in the information network; Represents data through the shortest path from the source node Send to Node The delay caused by the process.

[0057] In order to reflect the impact of information attacks on the timeliness of information networks, the relative transmission efficiency of information networks can be defined as for:

[0058] In the formula Represents the network transmission efficiency when the information network is not attacked.

[0059] The value range is .when When , it means that after the information network is attacked, the network transmission efficiency does not decrease; when When , it means that after the information network is attacked, the network transmission efficiency drops to zero.

[0060] Step 3.3: Load loss based on load importance

[0061] For general load loss indicators, simply adding up the load loss will result in large errors. In actual situations, the importance levels of various load nodes are different. In this case, it is necessary to set corresponding weights according to the importance levels of each load. In order to reflect the impact of information attacks on the physical power grid, the reduction of important loads is defined as an indicator for evaluating the severity of the consequences of the attack:

[0062] In the formula Represents load node Importance weight of load; Represents the active power of the power node before load reduction; Represents the active power of the power node after load reduction.

[0063] Step 3.4: Selection of power system vulnerability indicators includes: Table 1 Power system status evaluation indicators

[0064] Power system status assessment indicators: (1) Consequence hazard indicators: Instability node number indicator, node instability means that the voltage of a node exceeds the voltage stability range. Generally speaking, the voltage stability range in the power system is usually 90% to 110% of the rated voltage. If the voltage amplitude of a node exceeds this range, the node is called an unstable node. Assuming the attack Directly cause a change in a node or line while ignoring the intermediate process, attack The index of the number of unstable nodes caused by

[0065] In the formula For attack The number of unstable nodes caused, is the number of stable nodes in the power grid topology.

[0066] (2) Basic recovery time indicator: Instability node recovery time indicator, attack The recovery time index of the unstable node caused by , which reflects the power system's vulnerability to attacks The strength of the recovery capability is inversely proportional to the recovery time of the unstable node.

[0067] (3) Key indicators of operating status: Node connection status indicators, attacks The next node connection status indicator is,

[0068] In the formula is the number of connected nodes when there is no attack, For attack The number of downstream nodes.

[0069] Branch connection status indicators, attack The lower branch connection status indicator is,

[0070] In the formula is the number of all lines when there is no attack, For attack The number of valid lines.

[0071] Load fluctuation status indicator, attack The load fluctuation state index is:

[0072] In the formula is the base value of load power when there is no attack, For attack Real-time value of load power.

[0073] Step 3.5: Power system status assessment system; The analytic hierarchy process (AHP) is used to quantitatively evaluate the state of the power system. The hierarchical structure of the indicator system has been given above, and then the weight and score of each level indicator need to be calculated, and finally a comprehensive score is obtained based on the weight and score, which can reflect the state of the power system.

[0074] (1) Construct a weight matrix. Suppose the set of indicators at a certain level is , is the weight vector for obtaining the index , first construct the judgment matrix. Indicators and For upper level indicators The importance of . The value of is expressed in the AHP scale. The importance of each element at the same level to a criterion at the previous level is compared pairwise to obtain the judgment matrix , Considering that there may be inconsistencies when comparing the importance of indicators, it is also necessary to perform consistency checks on the constructed matrix. , If the consistency ratio , then the consistency of the judgment matrix can be considered acceptable, otherwise the judgment matrix needs to be modified.

[0075] (2) Weight calculation. We use the geometric mean method to calculate the weights. The weight vector of a certain layer is ,in , from which the importance ranking weight of the bottom-level indicators can be calculated.

[0076] (3) Calculation of index scores. The original scores of each index are calculated based on the actual situation of the evaluated system. The indexes are divided into positive indexes (the larger the index value, the better the power system status) and negative indexes (the larger the index value, the worse the power system status). All the above indicators are negative indicators, that is, the higher the score of the above indicators, the worse the power system status. The power system index score is recorded as ,in , is the raw score of each indicator, , It is the full score.

[0077] (4) Calculation of comprehensive score. The comprehensive score is the index score multiplied by the weight, expressed as The higher the comprehensive score, the better the power system status.

[0078] Step 3.6, weak links in the power system; The main weak links and vulnerable attack scenarios of the power system are as follows: (1) Attack scenarios on the power generation side. Local control: The control modules of the generator automatic excitation device and the speed regulator communicate with the control center via Ethernet and Modbus protocols. Local control data acquisition comes from the local area, and the attack range is small. It is necessary to mainly defend against malware that invades the substation LAN and attackers who invade the local communication network and tamper with the digital control logic or settings. Wide area control: Automatic generation control (AGC) is a means of secondary frequency regulation, which relies on the tie line flow and system frequency measurement values ​​collected by SCADA. Since the SCADA network coverage is relatively large, the risk of automatic generation control (AGC) being attacked is relatively high.

[0079] (2) Attack scenarios on the transmission side. The operations on the transmission side of the power system mainly use SCADA / EMS and relay protection equipment to monitor, evaluate and control the transmission network. Attacking the SCADA system includes illegal operations on the data exchanged in the network; attacking the EMS operation includes illegal modification of status assessment data, load forecast values, operator instructions, control actions, etc. WAMS provides synchronous acquisition and time-stamped voltage and current system-wide reference values ​​for power system protection, control or monitoring. The key to the reliable operation of this system is that the sampling clock of the entire system needs to be synchronized with the same time reference. The common attack is time-based attack.

[0080] (3) Attack scenarios on the distribution side. The AMI system involves communications between end users, control systems, and third-party organizations, and must ensure point-to-point information security. Malicious network attacks will undermine the confidentiality of smart meter data and the integrity and availability of remote switch control commands. In distribution automation, malicious attacks on equipment on monitoring feeders may lead to errors in automated operations such as remote switch opening and closing, capacitor access, and emergency response; malicious attacks on control logic or relay communication facilities in load management may cause unplanned distribution network feeder tripping, resulting in power outages. Distributed power sources include power generation devices and energy storage devices, which can serve as the main or emergency backup power source in the power grid to improve power quality and system operation stability. Malicious network attacks will undermine the confidentiality of user information of distributed power sources and the accuracy and timeliness of power grid measurement data. "A Review of Cyber ​​Attack Research in Power Information-Physical Fusion System Environment".

[0081] (4) Dispatching automation system. In commercial dispatching automation systems, the primary equipment of the power grid and its control ports are recorded in the form of data tables. According to the table structure of the dispatching system database, the remote control IP address and remote control number of each line and generator associated circuit breaker can be found, and then the control command and return signal can be sent according to the communication protocol to induce the circuit breaker to trip. It is even possible to select several key components with high betweenness or electrical betweenness rankings based on the power grid information and attack them one by one, so as to push the power grid into a self-organized critical state. Injection attack is to inject fictitious data into the dispatching system database to trigger the control system to malfunction, but its risk level is not as high as directly remotely tripping after invading the dispatching system.

[0082] The implementation of step 4 includes the following steps: The software UI interface is designed using the PyQt5 toolkit in Python. The codes for each model and specific algorithm are completed in Python. The performance of the software is tested using typical scenarios. After the test passes, the software is packaged.

[0083] The implementation of step 5 includes the following steps: 5.1 Establishing IEEE33 node distribution network model Figure 1 shown.

[0084] 5.2 Distribution network reconstruction modeling; The intelligent optimization algorithm using basic ring matrix coding usually uses an unordered solution space in dealing with the distribution network reconstruction problem. There are many local peaks in the solution space, which makes it difficult for the intelligent optimization algorithm to play its own advantages, is time-consuming and difficult to find the optimal solution. In view of the above problems, an ordered ring network coding method is proposed, and the reconstruction of the distribution network is solved based on the particle swarm algorithm. The particle swarm algorithm is used to solve the IEEE33 distribution network, which effectively reduces the system network loss and increases the minimum voltage in the system. It has been verified that this method is effective and feasible, the algorithm is simple, the speed is high, and the results are better.

[0085] Where: i represents the number of the branch; T is the set of closed branches in the distribution network; For the i The resistance of each branch; For the i The voltage effective value of the node at the end of each branch; and Respectively i The active and reactive power flowing through the end of a branch.

[0086]

[0087] The constraints are as follows: (1) Node voltage constraints:

[0088] (2) Branch capacity constraints

[0089] (3) Connected radial structure constraints:

[0090] Where: , , are the actual voltage of the end node of the i-th branch and its upper and lower limits respectively; , are the actual power flowing through the i-th branch and the maximum power allowed to flow through it; g is the network topology after the distribution network is reconstructed; G is the set of connected radial topology structures.

[0091] 5.3 Ordered ring network matrix coding; In the optimization process, the intelligent optimization algorithm generates new individuals based on the current optimal individual, and then searches for better individuals. However, when solving multi-peak function problems, there are many peaks (or valleys). When optimizing, it is possible to find a peak (or valley) and generate a new solution based on it, or affect the optimization of other individuals, which will cause the algorithm to fall into the local optimum. It is necessary to find a solution that is better than the peak (or valley) to jump out of the local optimum. However, in a single-peak function, there is no such situation of falling into the local optimum. And the more peaks a multi-peak function has, the easier it is for the intelligent optimization algorithm to fall into the local optimum when optimizing it. Therefore, when dealing with multi-peak problems, if the intelligent optimization algorithm can reorder its solution space to make it a single peak or reduce its peaks, the difficulty of searching can be greatly reduced, which can not only improve the quality of the solution, but also reduce the number of iterations.

[0092] (1) Label each branch in the system, 0 represents that the switch on the branch is open, and 1 represents that the switch on the branch is closed.

[0093] (2) Use graph theory to determine the basic loops. The number of loops is the same as the number of tie switches, which is 5 for a 33-node system.

[0094] (3) List the ring network matrix based on the basic rings. The number of rows in the ring network matrix is ​​the number of basic rings, and the number of columns is the number of branches of the ring with the largest number of branches. The empty spaces in other smaller rings that do not have so many branches are filled with 0.

[0095] 5.4 Distribution network reconstruction matlab code implementation includes code programming in matlab for IEEE33 node distribution network reconstruction.

[0096] Example This power system analysis case is specific to a radial network topology without distributed generation (DG) and tie switches. The model is based on the IEEE 33-node standard test system and is designed to analyze and optimize the operation of the network.

[0097] 1. Fault recovery in power systems; 1.1 Topological data processing; Extract and analyze the data of safe nodes and branches, and perform necessary processing, such as converting power units into standard units and extracting information such as transformer capacity. The network topology is defined by busdata and branchdata, where busdata contains information about all nodes and branchdata defines the connection relationship between nodes. Variables such as bus_line, j_bus_line,set_branch, i_bus_line_line, j_bus_line_line, set_branch_line, Num_sub, idx_sub are used to further process the network topology.

[0098] 1.2 Variable definition; The variables in the model are defined, including load recovery power, line status, virtual power, etc., and related constraints are established, including topology constraints, power balance constraints, line flow constraints, etc.

[0099] 1.3 Constraint establishment; 1.3.1 Power balance constraint (for each node i)

[0100] in is the set of nodes adjacent to node i, and is the active and reactive power generated by node i, and are the active and reactive losses of node i (usually negligible).

[0101] 1.3.2 Voltage Amplitude Constraints (For each node i );

[0102] in and are the minimum and maximum limits (per unit) of the voltage amplitude at node i.

[0103] 1.3.3 Branch flow constraints (for each branch i - j );

[0104] in is the complex power of branch ij (the sum of the squares of active and reactive power) is the maximum capacity limit of branch ij (per unit value).

[0105] 1.3.4 Radial topology constraints; The topology of the network must be radial, i.e. each node has only one parent node other than the root node, which is achieved by ensuring that no loops are formed in the network. In this model, this constraint is achieved by setting Num_root (number of root nodes) to 1 and limiting branch connections accordingly.

[0106] 1.3.5DG output upper limit constraint; Since this case assumes that there is no DG, the upper limit of DG output is set to 0, that is, the impact of distributed generation on the system is not considered.

[0107]

[0108] 1.3.6 Substation capacity constraints; If there are substations, the maximum capacity limit of the substation S_sub_max must be met.

[0109] 1.4 Objective function setting and solution The objective function aims to maximize the recovery of active load, that is, to minimize the total network loss, and then the optimization solver is used to solve it.

[0110]

[0111] in, is the power loss on branch i, and are the active and reactive power flows on this branch. is the resistance of branch ij (per unit value).

[0112] 1.5 Results Analysis The solution results are analyzed, including the calculation and output of indicators such as the system's load recovery level, important load recovery level, and load recovery power.

[0113] 2. Distribution network fault restoration simulation based on mixed integer linear programming (MILP); In power system analysis, especially after a distribution network fails due to natural disasters (such as hurricanes, earthquakes, etc.) or other reasons (such as equipment failure), it is necessary to restore power supply quickly and effectively. This code implements a MILP-based distribution network fault restoration simulation, which optimizes the restoration strategy to maximize the load restoration, and is used to set the parameters and initial data of the distribution network fault restoration problem.

[0114] 2.1 Analysis of distribution network after fault: When a fault occurs in a distribution network, the first task is to analyze the changes in the network's topology. This usually involves identifying which lines or nodes are disconnected due to the fault, resulting in changes in the network structure. By analyzing topological data such as branchdata and busdata, the scope of the fault impact can be determined. In addition, it is necessary to evaluate how these faults affect the network's power supply capacity, including power outages to loads and degradation of voltage quality.

[0115] 2.2 Island division and reconstruction: After a fault occurs, the distribution network may be split into multiple islands, that is, sub-networks that are no longer connected to the main grid. The process of islanding is to identify these independent sub-networks and determine the boundaries between them. This can usually be achieved by analyzing topological data such as set_branch.

[0116] Island reconstruction is to reconfigure the power sources and loads within or between islands after identifying them. This includes dispatching the output of distributed generation (DG) to meet the power needs of the islands; reconfiguring the loads to ensure that important loads are given priority power supply; and trying to restore or establish new line connections to enhance the connectivity and stability of the islands. These decisions are designed to restore as much power supply as possible and maximize the efficiency and reliability of power supply.

[0117] 2.3 Fault settings: Set the fault line and root node according to the number of islands after the fault.

[0118] 2.4 Optimizing Decisions: After the island is identified, a series of optimization decisions need to be made to restore power. These decisions include: (1) DG scheduling; Determine the output of each DG to maximize the power demand within the island.

[0119]

[0120] in It is the output of DG. and These are the upper and lower limits of DG output.

[0121] (2) Line flow constraints;

[0122] in, and are the active and reactive power flows from node i to node j, and is the power flow limitation of line ij.

[0123] (3) Load distribution; Redistribute the loads to ensure that important loads are powered first, taking into account the weight of the loads (such as w_load).

[0124] (4) Switch operation; Decide which line switches should be opened or closed to restore or establish new line connections and enhance the connectivity of the island. The switch state Sn is usually a binary variable, that is, Sn∈{0,1}.

[0125] (5) Network connectivity constraints; Ensuring that the network remains connected after failure recovery may require additional graph theory algorithms to ensure this.

[0126] Based on the above analysis, a series of constraints such as radial topology constraints, power balance constraints, and power flow constraints are established.

[0127] 2.5 Mixed Integer Linear Programming (MILP) Model; In order to solve the above optimization problem, it is usually modeled as a mixed integer linear programming (MILP) model. The model involves integer variables (such as the switching state of the line) and continuous variables (such as DG output, load distribution, etc.). The MILP model is able to handle this mixed variable problem and find the optimal solution or approximate optimal solution. In the model, various constraints (such as voltage limit, line flow limit, etc.) are expressed as linear inequalities or equations, while the objective function (such as maximizing the scope or efficiency of power restoration) is expressed as a linear function of the decision variables. By solving the MILP model, a series of optimal decision plans can be obtained to guide the fault recovery process of the distribution network.

[0128] In this embodiment, the goal of the mathematical model is usually to maximize the restoration of power supply to the power-lost load while considering various constraints. The following is the mathematical model formula: (1) To restore the power supply to the power-lost load to the greatest extent possible,

[0129] Where N is the total number of loads that have restored power supply, is the importance weight of load i, is the power value of load i.

[0130] (2) Minimize the number of switching operations,

[0131] Where M is the number of switching transitions, is the current state of switch n (0 or 1), is the initial state of switch n.

[0132] (3) Restoration of power supply considering load level,

[0133] in, is the power of the ith load, is the load status change (1 means power supply, 0 means power failure), is the load rating factor, and N is the number of power-off loads.

[0134] When the objective function is defined as maximizing the restoration of active load, that is, making the restored active load the largest, the MILP method is used to solve the defined constraints and objective function.

[0135] 3. Network reconstruction defense method; 3.1 Initialization and setting parameters include: 3.1.1 Initialize group parameters: (1) Define the number of particle swarms; (2) Determine the dimension of the search space, which refers to the number of operable tie switches; (3) Load the initial particle position from the IEEE33 node definition file and randomly generate the initial velocity; (4) Initialize the individual optimal position and global optimal position of the particle.

[0136] 3.1.2 This ring matrix: A basic ring matrix is ​​established, which defines the possible loops in the network and is used to subsequently determine which branches may be disconnected.

[0137] 3.1.3 Correlation Matrix: Create an association matrix, which is the network data loaded from the 33-node definition file and is used to describe the connection relationship between nodes and branches.

[0138] 3.2 The main cycle process includes: 3.2.1 Update weights and speeds: In each iteration, the weight coefficients are adjusted and the velocity of each particle is updated, which depends on the individual optimal and global optimal solutions, as well as a randomly generated learning factor.

[0139] 3.2.2 Update particle coordinates: The particle's position is updated based on its velocity, and after the position is updated, it is converted to the corresponding branch number in the basic ring matrix.

[0140] 3.2.3 Calculate fitness: (1) For each particle, the fitness of the grid is calculated based on its location (i.e., disconnected branch). This is achieved by modifying the distribution network configuration, performing power flow calculations, and calculating the total losses.

[0141] (2) Ensure that the network is still radial and that no loops are formed by checking the modified association matrix.

[0142] 3.2.4 Update individual optimal solution and global optimal solution: Update the individual optimal solution and the global optimal solution of each particle according to the new fitness value.

[0143] 3.3 Result Output and Visualization Calculate the grid loss and voltage distribution under the initial configuration, draw the voltage curve, and compare the changes before and after the network reconstruction.

[0144] Figure 2 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 2 As shown, the electronic device may include: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus. The processor may call the logic instructions in the memory to execute the power system attack and defense simulation method based on the abstract topology structure.

[0145] In addition, the logic instructions in the above-mentioned memory can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the 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 the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

[0146] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the power system attack and defense simulation method based on the abstract topology structure provided by the above-mentioned methods.

[0147] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the power system attack and defense simulation method based on an abstract topology structure provided by the above-mentioned methods.

[0148] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0149] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0150] Finally, 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 aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A power system attack and defense simulation method based on an abstract topology structure, characterized in that: The following steps are involved: Based on the steady-state power flow equation of the power grid, establish the information network model and the power information fusion system model; Build a network simulation platform for the power system, select the standard IEEE33 node system for power flow calculation and topology analysis, and verify the effectiveness of the model through case analysis; Considering the topology structure and the system operation status, select the information node loss ratio , the relative transmission efficiency of information networks As a vulnerability indicator of the information layer, a compliance loss based on load importance is defined. , select power system vulnerability indicators, and establish a power system status assessment system using the analytic hierarchy process based on power system security and stability theory; Distribution network reconfiguration modeling.

2. According to claim 1, the power system attack and defense simulation method based on abstract topology structure is characterized in that: The establishment of the information network model and the power information fusion system model based on the power grid steady-state power flow equation includes the following steps: 1.1) Establishing an information network model includes: The power information network is abstracted into a point set and edge sets The composition of the picture , the information transmission equipment is abstracted as the nodes of the graph, and the optical fiber transmission route is abstracted as the edges of the graph; based on the power information network, abstract modeling is performed to obtain a power information network model composed of different nodes and edges; 1.2) Modeling of power information fusion system: The power information fusion system is abstracted as a graph composed of point sets and edge sets, including different nodes and branches. The nodes are the nodes in the connection layer, backbone layer and core layer, and the branches are the branches in the power network and information network.

3. According to the power system attack and defense simulation method based on abstract topology structure according to claim 1, it is characterized in that: The construction of the network simulation platform of the power system, the selection of the standard IEEE33 node system for power flow calculation and topology analysis, and the verification of the effectiveness of the model through case analysis include the following steps: 2.1) Build a power system network simulation platform; Through platform virtualization, a logically exclusive computing environment is provided for nodes in the power system simulation network, and through network virtualization, logically independent network devices are provided for the construction of network links; 2.2) Perform power flow calculation and topology analysis; The constructed model is calculated using the MATLAB power flow calculation program, and the optimal network reconstruction method is obtained using the genetic algorithm.

4. According to claim 1, the power system attack and defense simulation method based on abstract topology structure is characterized in that: The selected information node loss ratio , the relative transmission efficiency of information networks As a vulnerability indicator of the information layer, a compliance loss based on load importance is defined. , select power system vulnerability indicators, based on the power system security and stability theory, use the analytic hierarchy process to establish a power system status assessment system including: 3.1) Information node loss ratio for, In the formula Represents the number of initial nodes of the information network; Represents the number of remaining nodes of the information node after the information system is attacked; 3.2) Select the relative transmission efficiency of the information network include: Information network transmission efficiency for, In the formula Represents the total number of nodes in the information network; Represents data through the shortest path from the source node Send to Node The delay caused by the process; Define the relative transmission efficiency of information networks for, In the formula Represents the network transmission efficiency when the information network is not attacked; The value range is ,when When , it means that after the information network is attacked, the network transmission efficiency does not decrease; when When , it means that after the information network is attacked, the network transmission efficiency drops to zero; 3.3) Define load loss based on load importance include: Define significant load reduction as load loss , In the formula Represents load node Importance weight of load; Represents the active power of the power node before load reduction; Represents the active power of the power node after load reduction; 3.4) Selected power system vulnerability indicators include: 3.4.1) Consequence hazard indicators; Set Attack Cause a change in a certain node or line while ignoring the intermediate process, attack The index of the number of unstable nodes caused by In the formula For attack The number of unstable nodes caused, is the number of stable nodes in the power grid topology; 3.4.2) The basic recovery time index, i.e. the recovery time index of the unstable node, is: attack The recovery time index of the unstable node caused by , which reflects the power system's vulnerability to attacks The strength of the recovery capability is inversely proportional to the recovery time of the unstable node; 3.4.3) Key indicators of operating status include: Node connection status indicators; Attack The next node connection status indicator is, In the formula is the number of connected nodes when there is no attack, For attack Number of downstream nodes; Branch connection status indicator; attack The lower branch connection status indicator is, In the formula is the number of all lines when there is no attack, For attack The number of valid lines. Load fluctuation status indicator; attack The load fluctuation state index is: In the formula is the base value of load power when there is no attack, For attack Real-time value of load power; 3.5) Establishment of power system status assessment system including: The analytic hierarchy process is used to quantitatively evaluate the power system status, calculate the weight and score of each level indicator, and obtain a comprehensive score based on the weight and score. The comprehensive score is used to reflect the power system status. The specific steps are as follows: 3.5.1) Construct a weight matrix; the set of indicators at a certain level is , is the weight vector for obtaining the index , construct the judgment matrix; use Indicators and For upper level indicators The importance of ; The value of is expressed by the scale of the hierarchical analysis method; the importance of each element at the same level with respect to a criterion at the previous level is compared pairwise to obtain the judgment matrix , perform consistency check on the constructed judgment matrix. If the consistency ratio , then the consistency of the judgment matrix is ​​acceptable, otherwise the judgment matrix needs to be modified; 3.5.2) Weight calculation: The geometric mean method is used to calculate the weight. The weight vector of a certain layer is ,in, , thereby calculating the importance ranking weight of the lowest level indicators; 3.5.3) Calculation of index scores; the index scores of power systems are recorded as ,in , is the raw score of each indicator, , is the full score; 3.5.4) Calculation of comprehensive score; the comprehensive score is the indicator score multiplied by the weight, recorded as ; The higher the comprehensive score, the better the power system status; 3.6) Weak links in the power system include: 3.6.1) Attack scenarios on the power generation side; 3.6.2) Attack scenarios on the transmission side; 3.6.3) Attack scenario on the power distribution side; 3.6.4) Dispatching automation system.

5. The power system attack and defense simulation method based on abstract topology structure according to claim 1 is characterized in that: The distribution network reconstruction modeling includes: firstly establishing an IEEE33 node distribution network model, establishing a distribution network reconstruction model and an ordered ring network matrix coding according to a particle swarm algorithm, and implementing distribution network reconstruction code programming in matlab; the specific steps are as follows: 4.1) Distribution network reconstruction modeling includes: The ordered ring network coding method is adopted, and the reconstruction of the distribution network is solved based on the particle swarm algorithm; the particle swarm algorithm is used to solve the IEEE33 distribution network formula as follows: In the formula, i represents the number of the branch; T is the set of closed branches in the distribution network; For the i The resistance of each branch; For the i The voltage effective value of the node at the end of each branch; and Respectively i Active and reactive power flowing through the end of each branch; The constraints are as follows: (1) Node voltage constraints: (2) Branch capacity constraints: (3) Connected radial structure constraints: Where: , , are the actual voltage of the end node of the i-th branch and its upper and lower limits respectively; , are the actual power flowing through the i-th branch and the maximum power allowed to flow through it; g is the network topology after the distribution network is reconstructed; G is the set of connected radial topology structures; 4.2) Ordered ring network matrix coding includes: 4.2.1) Label each branch in the system, 0 represents the switch on the branch is open, 1 represents the switch on the branch is closed; 4.2.2) Use graph theory to determine the basic loops. The number of loops is the same as the number of tie switches. 4.2.3) List the ring network matrix based on the basic rings. The number of rows in the ring network matrix is ​​the number of basic rings, the number of columns is the number of branches of the ring with the largest number of branches, and the other rings are filled with 0.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the power system attack and defense simulation method based on the abstract topology structure as described in any one of claims 1 to 5 is implemented.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the power system attack and defense simulation method based on an abstract topology structure as claimed in any one of claims 1 to 5 is implemented.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the power system attack and defense simulation method based on an abstract topology structure as claimed in any one of claims 1 to 5 is implemented.

Citation Information

Patent Citations

  • Electric power system vulnerability assessment method

    CN104156769A

  • A distribution network reconfiguration method based on immune binary particle swarm optimization algorithm

    CN109217284A

  • Intelligent power grid information physical system vulnerability assessment method based on complex network

    CN116011885A

  • Optimal load transfer method for access of new energy and flexible interconnection equipment to power distribution system

    CN117833253A

Cited By

  • Novel power distribution network fault recovery method

    CN119482463A

  • Method for evaluating post-earthquake toughness of multi-scale power system

    CN119761646A

  • Power network key circle detection method based on simulated annealing algorithm

    CN120257865A

  • Network anti-attack performance evaluation method and device, computer equipment and storage medium

    CN120321047A