Event-power grid-lifeline-user-oriented disaster influence full-chain joint simulation method

Abstract

The invention relates to an event-power grid-lifeline-user-oriented disaster influence full-chain joint simulation method, and aims to solve the problems that composite extreme events cannot be covered, coupling modeling is inaccurate and a unified full-chain simulation framework is lacked in the prior art. According to the method, extreme event scenes such as natural disasters and network attacks are generated, a multi-granularity lifeline model is constructed, joint simulation is carried out in two stages (two-way coupling system iteration simulation and one-way influence system section simulation), and eight types of important users are synchronously modeled by combining time-event dual-drive interaction. The full-chain failure process can be accurately simulated, power grid fault evolution, lifeline function loss and user service availability are output, and support is provided for urban lifeline toughness evaluation and emergency decision making.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power grid planning, in particular to a disaster impact full-chain joint simulation method for event-power grid-life line-user. BACKGROUND

[0002] With the acceleration of urbanization, the urban life line centered on the urban power grid, covering the natural gas pipeline network, water supply network, road traffic network, public communication network and rail transit system, has become the key to guarantee the operation and public safety of the city. However, extreme events (including natural disasters, network / physical attacks, major accidents and compound chain events) not only directly damage power grid equipment, but also cause cascading failures through system coupling, impacting important users such as hospitals and transportation hubs, causing economic losses and safety risks. Therefore, building an extreme event-urban power grid-urban life line-important user full-chain impact analysis model is the core demand to improve system resilience and support consequence assessment.

[0003] However, the existing technology has obvious defects: first, it ignores compound events and only focuses on single disasters, not covering network attacks combined with physical attacks, typhoons combined with rainstorms and other chain events, so the fault scenarios deviate from the actual situation; second, the coupling modeling is fragmented, without considering bidirectional coupling (such as power grid-gas network mutual feedback), the dynamic process is oversimplified (such as gas network using a steady-state model), and the time scales of each system are not coordinated; third, there is no unified joint simulation framework, tools are independent, lack of automatic data interaction and event-time driving fusion, and cannot simulate the full-chain evolution; fourth, important user modeling is not complete, without considering multi-resource dependency, and the classification coverage is insufficient.

[0004] In summary, the existing technology cannot meet the full-chain analysis demand, and new methods are urgently needed to improve the resilience of urban life lines. SUMMARY

[0005] The purpose of the present application is to overcome the defects of the existing technology and provide a disaster impact full-chain joint simulation method for event-power grid-life line-user, to solve or partially solve the problem that the existing scheme cannot comprehensively cover compound chain extreme events, accurately depict the coupling relationship and dynamic interaction between urban life line systems, and thus cannot effectively simulate the full-chain coupling failure process to support urban life line system resilience assessment and emergency decision-making.

[0006] The purpose of the present application can be achieved by the following technical solutions: In one aspect of the present application, a disaster impact full-chain joint simulation method for event-power grid-life line-user is provided, comprising: S1: generating an extreme event scenario, analyzing the action mode of the extreme event on the urban power grid, and determining the initial fault parameters of the power grid; S2: Model the urban power grid, natural gas pipeline network, road traffic network, water supply network, public communication network and rail transit system respectively to obtain the dynamic model of the urban lifeline system; S3: Model multiple coupled components among urban lifeline systems to obtain a parameter correlation model; S4: Model the resource dependencies of critical users on the lifeline system; S5: In the first stage, iterative simulations are performed on the power grid-natural gas pipeline network and the power grid-road traffic network with bidirectional coupling characteristics, and the coupling point information is exchanged at a fixed step size. In the second stage, cross-sectional simulations are performed on the water supply network and public communication network that are affected by the power grid in one direction, and the functional loss is calculated based on the fault parameters to realize the interaction of time-driven and event-driven approaches. S6: Output joint simulation results, including the urban power grid fault evolution process, the functional loss rate of each lifeline system, and the service availability indicators of important users.

[0007] As a preferred technical solution, the modeling process of the urban lifeline system dynamics model includes: A steady-state model of AC optimal power flow is obtained by modeling the urban power grid; a simplified dynamic model is obtained by modeling the natural gas pipeline network; a hybrid model combining a macroscopic user equilibrium traffic assignment model and a microscopic vehicle following model is obtained by modeling the road traffic network; a steady-state hydraulic model is obtained by modeling the water supply network; a directed weighted graph theory model is obtained by modeling the public communication network; and a dual-subject model of train and passenger is performed on the rail transit system; thus, a dynamic model of the urban lifeline system is obtained.

[0008] As a preferred technical solution, the hybrid model combining the macroscopic user equilibrium traffic assignment model and the microscopic vehicle following model is modeled as follows: Macro-level user equilibrium traffic allocation model: , , , , , In the above formula, For road section a Traffic flow For traffic flow aggregation, Represents a cost or time function. Indicates from the source node r To the target node s Total demand Representing a path l Traffic, For indicator functions, when the path contains road segmentsa 1 if t is 1, otherwise 0, denotes a path from a source node r to a target node s , denotes a constraint condition; Microscopic car-following model: , , , In the above formula, denotes the safety distance speed at time , denotes the front vehicle speed at time , denotes the vehicle spacing at time , is the braking reaction time, is the following reaction time, denotes the decision speed at time , denotes the maximum speed limit of the vehicle, denotes the current vehicle speed at time , denotes the vehicle acceleration related to the current speed, denotes the true speed of the vehicle, is the time slice, is the desired spacing at time , is the spatial position coordinate of the current vehicle at time

[0009] As a preferred technical solution, the constraint condition of the hybrid model modeling includes a cell traffic flow balance constraint, a constraint that the traffic flow of a resource vehicle from a cell to an adjacent cell is not greater than the traffic flow of the cell, a constraint that the total traffic flow of vehicles from the cell to the adjacent cell is not greater than the upper limit of the traffic flow transfer of the cell, a constraint that the total traffic flow of vehicles from the adjacent cell to the cell is not greater than the upper limit of the traffic flow transfer of the cell, a constraint that when the cell is in traffic congestion, the total inflow of the cell does not exceed the remaining capacity of the cell, and a constraint of the initial and terminal traffic flow states in each cell.

[0010] As a preferred technical solution, the step S5 includes: S51: initialize simulation parameters, set the simulation duration , power grid scheduling period and natural gas pipe network simulation step size , wherein is an integer; ​S52: if the current time , execute spatiotemporal discretization calculation time system state, update to ; S53: in response to detecting an event, performing event-driven interaction, power grid and gas network exchange gas turbine operating state and power consumption of electric drive facility, power grid performs AC-OPF scheduling and feeds back gas consumption of gas turbine, wherein the event includes initial fault, gas network cascading failure and power grid cascading failure; S54: if the current time , execute time-driven interaction, gas network transmits power consumption of electric drive facility to power grid, and power grid transmits gas consumption to gas network after performing AC-OPF scheduling; S55: after completing the first-stage bidirectional coupling simulation, inputting power grid fault parameters to water supply network and communication network, calculating water supply loss and communication node failure number, outputting life line system function loss, and completing the second-stage simulation.

[0011] As a preferred technical solution, the resource dependency relationship of important users to the life line system is modeled through a function function model or a subject modeling.

[0012] As a preferred technical solution, the important users include command type, communication type, news media type, data and financial center type, water and heat and gas supply type, transportation type, medical and health type and important venue type.

[0013] As a preferred technical solution, the coupling elements include electric drive compressor, electric drive gas source and gas turbine of power-natural gas network, traffic signal lamp of power-road traffic network, water pump of power-water supply network, base station / switch of power-public communication network and train of power-rail transit system.

[0014] As a preferred technical solution, the extreme event scenario includes natural disaster, network attack, physical attack, major unexpected accident and composite chain extreme event.

[0015] Another aspect of the application provides a disaster impact whole-chain joint simulation system for event-power grid-life line-user, which is used for realizing the aforementioned disaster impact whole-chain joint simulation method for event-power grid-life line-user, and the system comprises: An extreme event fault scenario generation module is used for generating an extreme event scenario, analyzing the action mode of the extreme event on the urban power grid, and determining power grid initial fault parameters. An urban life line system modeling module is used for modeling the urban power grid, natural gas pipeline network, road traffic network, water supply network, public communication network and rail transit system respectively, and obtaining an urban life line system dynamics model. a coupling element modeling module configured to model a plurality of coupling elements between urban lifelines to obtain a parameter correlation model; a resource dependency modeling module configured to model resource dependency relationships of important users to lifelines; an urban lifeline simulation module configured to perform iterative simulation on power grid-natural gas pipeline network and power grid-road traffic network with bidirectional coupling characteristics in a first stage, exchange coupling point information at a fixed step, perform section simulation on water supply network and public communication network influenced by power grid in a single direction in a second stage, calculate function loss based on fault parameters, realize interaction of time driving and event driving, and output joint simulation results, including urban power grid fault evolution process, function loss rate of each lifeline, and service availability index of important users.

[0016] Compared with the prior art, the present application has at least one of the following beneficial effects: (1) High simulation accuracy: To solve the problems of the current scheme, such as insufficient real-time interaction of coupling point information, lack of accurate input for function loss calculation, large deviation of results, and incoherent system state synchronization, the present application performs simulation in two stages, iteratively simulates power grid-natural gas pipeline network and power grid-road traffic network with bidirectional coupling characteristics in a first stage, exchanges coupling point information at a fixed step, performs section simulation on water supply network and public communication network influenced by power grid in a single direction in a second stage, calculates function loss based on fault parameters, realizes interaction of time driving and event driving, thereby accurately capturing the bidirectional cascading fault evolution process, providing accurate data support for function loss evaluation of lifelines through section simulation of actual fault parameters of power grid, overcoming the simulation fracture problem through the two-stage combined double-driving method, realizing full-chain automatic simulation from extreme event input to important user impact output, and providing reliable simulation basis for urban lifeline system resilience evaluation and emergency decision-making.

[0017] (2) Generation of composite fault scenarios close to actual disasters: To solve the problem of the prior art that focuses on a single type of extreme event, such as a single typhoon or a single network attack, without considering composite chain generation scenarios such as "network attack weakening defense followed by physical attack, typhoon superimposed with heavy rain causing secondary disasters, resulting in large deviation of the generated fault scenarios from actual disasters, the present application generates extreme event scenarios, analyzes the action mode of extreme events on urban power grids, determines the initial fault parameters of power grids, covers natural disasters, network attacks, physical attacks, major accidents, and composite chain generation extreme event disasters, analyzes the action mode of each type of event on power grids, clearly defines the evolution logic of composite chain generation events, forms an extreme event fault scenario generation method covering all types, and thereby generates composite fault scenarios close to actual disasters, completely covering the influence path of extreme events on power grids, and providing accurate initial fault input for subsequent full-chain simulation.

[0018] (3) The accuracy of the fault cross-system propagation simulation is high: in view of the fact that the prior art models a lifeline system alone, ignores the bidirectional coupling between systems, and greatly simplifies the key dynamic process, and the time scales of the systems are not coordinated, which leads to a large deviation in the timing prediction of the fault cross-system propagation, the present application constructs a multi-granularity lifeline system dynamics model, establishes a coupling element parameter correlation model, and designs a hierarchical simulation based on the differences in the time scales of the systems, so as to accurately depict the bidirectional coupling effect and dynamic evolution process between lifeline systems, and improve the accuracy of the fault cross-system propagation simulation. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 A flowchart of the disaster impact whole-chain joint simulation method for an event-power grid-lifeline-user in the embodiment; Figure 2 A schematic diagram of whole-chain key infrastructure modeling in the embodiment; Figure 3 A schematic diagram of multi-stage joint simulation in the embodiment; Figure 4 A schematic diagram of a power system cascading failure model in the embodiment; Figure 5 A schematic diagram of a fault propagation path between urban lifeline systems; Figure 6 A schematic diagram of a disaster impact whole-chain joint simulation system for an event-power grid-lifeline-user in the embodiment. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative labor should fall within the scope of protection of the present application.

[0021] Embodiment 1 In view of the problems of the prior art described above, the present embodiment provides a disaster impact whole-chain joint simulation method for an event-power grid-lifeline-user, as shown in Figure 1 The method comprises the following steps: S1: generating an extreme event scenario, analyzing the action mode of the extreme event on the urban power grid, and determining the initial fault parameters of the power grid.

[0022] Disaster scenarios are the basis of resilience assessment. The extreme events modeled in this step include four types of extreme disasters, i.e. natural disasters, cyber attacks, physical attacks and major accidents, and compound chain extreme events. The impact mode of these extreme events on the power system is analyzed to generate typical fault scenarios, which can be used for modeling and simulation of urban lifeline systems and assessment of the resilience level of urban power grids.

[0023] The action mode of extreme events is shown in Table 1.

[0024] Table 1 Action mode of extreme events

[0025] S2: Model the urban power grid, natural gas pipeline network, road traffic network, water supply network, public communication network and rail transit system respectively to obtain the dynamic model of the urban lifeline system.

[0026] This step builds a dynamic model of the urban lifeline coupling system with the power grid as the core and including the natural gas pipeline network, water supply network, public communication network and road traffic network, which dynamically reflects the impact of a large-scale power outage caused by extreme disasters on the functions of urban lifelines. The system topology and coupling relationship between each subject are analyzed, and the fault propagation path between urban lifelines under extreme disasters is shown in Fig. 1. Figure 5

[0027] The modeling process of each part is described as follows: (1) Power grid model The model is an existing AC optimal power flow model, and the constraints of the model include: real-time active and reactive power balance constraints of nodes; upper and lower limit constraints of branch flow in and out; reference node phase angle equalization constraint; upper and lower limit constraints of bus voltage size and generator injection.

[0028] (2) Simplified dynamic model of natural gas pipeline network The model is as follows:

[0029] In the formula, ρ is the density of gas; t is the time variable; v is the flow rate of gas; x is the variable along the pipeline axis; p is the pressure of gas in the pipeline; f is the hydraulic friction coefficient; D is the diameter of the pipeline; g is the acceleration of gravity; cp is the constant volume specific heat capacity; T is the thermodynamic temperature; and ​​​​​​​​​​​​​Cp, R, T A is the cross-sectional area of the pipe; is the horizontal inclination of the pipe axis; is the inclination of the pipe; is the energy exchange term.

[0030] Based on the above, the following simplifications are made: 1. In the urban power-gas coupled system, it is assumed that the urban temperature change is small, and the temperature change of the gas inside due to heat conduction between the pipe wall and the soil is relatively slow and can be ignored. Therefore, it can be considered that the gas flow is isothermal flow, so the model can not consider the energy equation, and the gas state equation can be simplified as:

[0031] where, is the isothermal wave speed.

[0032] 2. Numerical experiment simulation shows that ignoring the convection term mainly affects the fluctuation amplitude and damping before the flow reaches the steady-state condition, and ignoring the inertia term mainly affects the inability to capture the physical oscillation process when the boundary condition changes suddenly, and has less effect on the unit pressure recovery or decline time. Therefore, the model can ignore the inertia term and the convection term 3. In relatively flat urban areas, the terrain is not high, so the effect of the gravity term is also ignored.

[0033] The simplified, the simplified dynamic model of natural gas suitable for fault evolution analysis of urban power grid-natural gas network is:

[0034] where, is the mass flow rate.

[0035] (3) Road traffic network model By constructing a cellular transmission model, the simulation analysis of road delay caused by traffic light failure due to power failure is realized: (2-17)

[0036]

[0037] (2-19)

[0038]

[0039]

[0040]

[0041]

[0042] In the formula, for Time Cell It has a built-in charging requirement ( ) and vehicles without charging needs ( Traffic flow; for Time by cell j Flowing out to adjacent cells Traffic flow; For cells upstream cells; A set of cells; For time sets; For cells Downstream cells; A cell within a time period Traffic flow limit; For cells The congestion coefficient; For cells The maximum total traffic volume limit; For cells Internal resources The initial traffic flow distribution; For cells To cell resources The initial transfer traffic flow; It is the set of connections between cells; The set of cells in the starting region; To reach the set of regional cells.

[0043] Equation (2-17) represents the traffic flow balance for each cell. Equation (2-18) represents the resource... Vehicles from cells Traffic flow to an adjacent cell is no greater than the traffic flow to that cell. Equation (2-19) represents the flow of vehicles from a cell to an adjacent cell. The total traffic flow to adjacent cells does not exceed the traffic flow transfer limit of that cell. Equation (2-20) represents the traffic flow from adjacent cells to cells. The total traffic flow does not exceed the traffic flow transfer limit of this cell. Equation (2-21) indicates that when traffic congestion occurs in this cell, the cell... The total inflow does not exceed the cell The remaining capacity. Equations (2-22) and (2-23) represent the initial and final traffic flow states in each cell, respectively.

[0044] The traffic network of the present embodiment adopts a modeling method combining macro and micro. When there is no event triggering, the dynamic of the traffic network does not need to be described in detail, and a user equilibrium traffic assignment model is adopted for macro modeling to provide initial values for the dynamic process modeling after the event triggering. When the event triggers (such as a traffic light failure), the behavior of each vehicle in the intersection needs to be described in detail, and the traffic delay needs to be analyzed. Therefore, after the event triggers, a micro vehicle following model is adopted for modeling.

[0045] User equilibrium traffic assignment model (macro):

[0046] In the formula, denotes a cost or time function; denotes the total demand from the source node r to the target node s ; denotes the flow of the path l ; is an indicator function, which is 1 when the path contains the road segment a ; is a road segment set; is a path type, such as a toll / free path, a highway / ordinary path; is the total flow of the road segment .

[0047] Vehicle following model (micro):

[0048] In the formula, denotes the safe distance speed at time ; denotes the speed of the preceding vehicle at time ; denotes the vehicle spacing at time ; is the desired spacing at time ; denotes the braking time; denotes the driver reaction time; denotes the maximum speed limit of the vehicle; is the simulation time step; . denotes the current vehicle speed at time ; denotes the vehicle acceleration related to the current speed; denotes the speed after the vehicle challenges; is the target speed.

[0049] (4) Water supply network model The modeling is:

[0050] wherein, , are pipe, water pump branch set respectively; is water flow from node i to node j ; , are water head of node i , node j respectively; is water flow from node k to node i ; is external supply water amount of node i ; is actual consumption water amount of node i ; is water head characteristic coefficient of water pump branch.

[0051] (5) Rail transit system model The subway station is modeled by using the agent modeling method, each passenger and each train is set as an agent, and different behavior criteria and parameters are given, so that the dynamic changes of the train and the passenger under the influence of the power failure accident can be simulated, and the passenger flow distribution in the subway station, the train operation state and the number of people trapped under extreme events can be more intuitively reflected. The specific modeling method is as follows: 1) Train operation dynamic change

[0052] wherein, , are the positions of the train at the next moment and the current moment respectively; , are the speeds of the train at the next moment and the current moment respectively; is the acceleration of the train; , are the upper and lower limits of the train acceleration; is the upper limit of the speed of the train; is the time interval.

[0053] 2) Dynamic change of passenger behavior

[0054] wherein, is used to represent the behavior state of the passenger in the subway station, and a group of variables are used to describe; is the position of the passenger at the moment ; is the speed of the passenger at the moment. For time intervals.

[0055] 3) Train-passenger interaction behavior The interaction between the subway intelligent agent and the passenger intelligent agent mainly occurs during the process of train arrival at the station and passengers getting on and off the train.

[0056]

[0057] In the formula, This refers to the actual number of passengers who boarded the bus. for t The total number of passengers waiting to board this train at the current station; Indicates the maximum number of passengers allowed on the train; for Number of passengers on the train at any given time; for Passengers on the train at any given time; for t The actual number of passengers disembarking at the current station on the train; The total number of passengers scheduled to disembark; This represents the number of passengers waiting to disembark at the current station on the train.

[0058] (6) Public communication network model To consider the impact of the power grid on the public communication network, and based on the characteristics of the communication network, the impact of power outages on both wireless and wired networks is considered, leading to the establishment of a graph theory model for the public communication network. When analyzing the impact of power outages on the communication network, it is constructed as a directed weighted graph. This is because data flow and signal transmission in communication networks are often directional, especially when considering the interaction between devices such as data centers, routers, and repeaters. In the directed graph model, nodes represent communication devices or network nodes, such as base stations, switches, and servers, while directed edges represent the direction of data flow, i.e., communication links from one node to another. In this model, each edge not only has a direction but also a weight, which represents different characteristics of the communication link, such as bandwidth, latency, reliability, or capacity. Specifically, this is represented as follows:

[0059] In the formula, It is a directed weighted graph; Nodes in a network are represented; base stations, switches, etc., are all modeled as nodes. Indicates the first n One communication node; Represents the edges in the network, indicating the relationships between base stations, switches, users, etc. Represents a directed edge The weight parameters.

[0060] In summary, the coupling dynamics modeling framework for the whole chain critical infrastructure is shown in Figure 2

[0061] S3: Model the multiple coupling elements between urban lifeline systems to obtain the parameter correlation model.

[0062] (1) Power-natural gas network The coupling elements of power and natural gas network mainly consider electric-driven compressor, electric-driven power supply and gas turbine unit.

[0063] Electric-driven compressor: power consumption of compressor The calculation is as follows:

[0064] In the formula, represents the fraction of the total driving power provided by the electric drive; is the isentropic exponent of natural gas; is the gas constant of natural gas; represents the density of natural gas under reference conditions; is the compressibility factor; is the product of the adiabatic efficiency and the driving efficiency of the compressor; is the mass flow rate of the compressor; is the pressure at the outlet of the compressor; is the pressure at the inlet of the compressor; is the thermodynamic temperature of the inlet natural gas of the compressor.

[0065] Electric-driven gas source: electric-driven gas source needs reliable power supply to maintain normal operation, in general, the natural gas output of electric-driven gas source is closely related to the power consumption , which is expressed as:

[0066] In the formula, represents the conversion coefficient of electric-driven gas source.

[0067] Gas turbine unit: in the coupling analysis, the relationship between the required gas mass flow rate and the output power of the gas turbine unit is as follows. It is worth noting that the normal operation of the gas turbine unit needs to meet a certain pressure threshold, when the pressure is lower than the set value, the gas turbine unit will exit operation.

[0068]

[0069] In the formula, , and are parameters determined by the heat consumption rate curve of the gas turbine, ​Output power of the gas turbine unit.

[0070] (2) Power-road traffic network The coupling element of power and road traffic network mainly considers traffic lights, and the traffic delay model is established based on queuing theory and signal control theory.

[0071]

[0072] In the formula, and respectively represent the delay under power outage and normal condition; is the service time (i.e. green light time) of the intersection in direction i under normal condition; is the traffic density of the intersection in direction i , represents the service rate; is the number of vehicles arriving at the intersection in direction i per unit time; is the total red light time in all directions. is the additional delay factor of the intersection in direction i , considering the confusion and mismanagement caused by power outage. is the traffic density under power outage, may be reduced due to power outage.

[0073] (3) Power-water supply network The coupling element of power and water supply network mainly considers water pump, which consumes power and the flow consumed by water flow The relationship is:

[0074] In the formula, is the water density; is the fluid gravity coefficient; is the water pump efficiency; and are constants.

[0075] (4) Power-communication network Since the communication network adopts a graph theory-based model, the coupling elements (base stations, mobile switches, gateway routers, public communication centers) are modeled as nodes in graph theory, and the graph theory-based cascading failure model is used to analyze the consequences of power outage on the communication network.

[0076] (5) Power-rail transit system Since the rail transit system adopts the main body modeling method, the coupling element (train) is modeled as a main body with multiple parameters, and the dynamic evolution mechanism of the main body model is used to analyze the consequences of power failure on the rail transit system.

[0077] S4: Model the resource dependency relationship of important users to the lifeline system.

[0078] In this step, according to the standard of "Technical Specification for Power Supply and Emergency Power Supply Configuration of Important Power Users", the important users of the power system are classified by comprehensively considering the important position of the users in social and economic life and the widespread impact that power failure may cause. Eight categories including command, communication, news media, data and financial center, water, heat and gas supply, transportation, medical and health, and important venue are determined, totaling thirty key power users.

[0079] According to the different characteristics of each user and the impact of power failure on them, different modeling methods can be used to analyze their urban public safety characteristics, including the function function modeling method and the agent modeling method. The basic resources relied on by important users are used as modeling input, and the corresponding output that can reflect the multi-level function of important users is calculated using the above methods.

[0080] Optionally, the modeling calculation method corresponding to each important user includes a function function model, a direct quantitative calculation function, and an intelligent agent modeling.

[0081] (1) Important user modeling based on function function According to the service characteristics of important users such as banks and hospitals, the modeling method of function function is used to calculate the service availability. The user function is regarded as a function module, and the service level is defined through input and output. The internal structure or operation mechanism does not need to be considered in detail. Each function function in the model represents an operation unit of the system, has a certain input and output, and a conversion mechanism to convert input to output. The functions are connected through the flow of data or signals. This model represents the service availability of important users based on the principle of the wooden barrel, which is as follows:

[0082] In the formula, is the service availability of the system, is the input quantity that affects the function of the system, which reflects that the overall performance of the system is limited by the weakest part (2) Important user modeling based on agent modeling Agent-based modeling method is a method that can simulate the individual or collective dynamic behavior and their interaction in complex systems. In the agent-based modeling method, each agent is set with specific parameters and behavior rules, so that the agents can act autonomously in the simulation environment and interact with other agents or the environment itself.

[0083] For example, consider the modeling of urban rail transit passenger stations, which are important users in the power system. Using the agent-based modeling method, the system function level is quantitatively calculated, and the direct harm to the urban rail transit system caused by power grid failures, such as train emergency braking and passenger entrapment, is analyzed, which affects the urban public safety events. Then, the urban public safety assessment under extreme event scenarios is completed.

[0084] S5: In the first stage, the power grid-natural gas pipeline network and the power grid-road traffic network with bidirectional coupling characteristics are iteratively simulated, and the coupling point information is exchanged at a fixed step. In the second stage, the water supply network and the public communication network affected by the power grid in one direction are simulated at the section level. Based on the fault parameter calculation function loss, the combination of time-driven and event-driven interaction is realized.

[0085] This step proposes a joint simulation method suitable for "extreme event-urban power grid-urban lifeline-important user" whole-chain influence analysis. According to the influence mode of multiple types of disasters, typical disaster scenarios under different disasters are generated, and the fault evolution process after the disaster is further analyzed, which intuitively reflects the dynamic propagation process. Since the power and natural gas networks, and the power and traffic networks have bidirectional coupling characteristics, continuous iteration and interaction of information is required at the coupling level of the power grid and natural gas. Other networks such as communication networks and water networks are affected by the power grid in one direction, and do not involve the interaction and propagation of faults. When necessary, interact and calculate the loss of consequences.

[0086] (1) Joint simulation process The dynamic constants of urban lifeline systems range from a few milliseconds to a few hours, i.e., the transmission of two types of energy occurs within different time scales. In order to accurately describe the influence of one system on another system and analyze the fault evolution process, whole-chain joint simulation needs to be realized in a unified framework.

[0087] This step establishes a multi-level joint simulation framework according to the coupling characteristics of the power network and other different networks. The conceptual diagram is as follows Figure 3The disaster scenario generated by the extreme event is taken as the input of the urban lifeline simulation. In the first stage of the urban lifeline simulation, the network with bidirectional coupling characteristics is simulated, which needs to interact information at a specific time. In this stage simulation, a joint simulation method combining time-driven and event-driven is proposed, each subsystem independently carries out simulation, exchanges coupling point information at a fixed step, and interpolates important events such as coupling element failure to accurately simulate the fault cascading process. The second stage is to analyze the network affected by the power grid in one direction, accept the coupling element failure and recovery time of the water network and communication network output by the first stage simulation, and calculate the service function loss of the water network and communication network. Among them, the water supply network considers the influence of water storage facilities, and the communication network considers the influence of standby power sources such as base stations, and can be analyzed by section at a specified time interval. Finally, the important user simulation module is entered, and the function service level is output.

[0088] Under this framework, joint simulation takes into account both time and space dimensions. In the time dimension, due to the different time scales of different energy systems, the differences in dynamic characteristics between systems are considered in joint simulation. The power system can reach steady state on the time scale of milliseconds, so a steady-state model is used. Compared with the power system, the natural gas system has slow dynamic characteristics, with a time scale of minutes or even hours, so a dynamic model is needed to reflect the slow dynamic change process of each variable. In the fault phase, the power grid outage mainly affects the traffic light facilities of the road traffic network, and the deployment of mobile emergency resources in the recovery phase will affect the power restoration process. Since the running state of the vehicle needs to be considered in modeling, a micro-based vehicle following model is established. Other networks focus on analyzing the impact of power grid failure on their respective functions. According to the modeling idea in step S3, the water supply network uses a steady-state hydraulic model for modeling, and the communication network uses a graph model for modeling. In the spatial dimension, this step considers the cross-space interaction and propagation process of faults through coupling elements, the fault recovery process, and the cross-space influence process of urban lifeline functions.

[0089] In the first level simulation, a joint simulation method combining time-driven and event-driven is proposed, each subsystem independently carries out simulation, exchanges coupling point information at a fixed step, and interpolates important events such as coupling element failure to accurately simulate the fault cascading process.

[0090] This step focuses on the fault evolution process after the initial fault caused by an extreme disaster. The joint simulation method is shown in Table 2. At the beginning of joint simulation, the power system and the natural gas system should interact information to determine the initial state of the coupled system and provide initial values for simulation. The power system updates its state according to its dispatching period every A steady state analysis is performed. For the natural gas system, since it cannot reach the next steady state as quickly as the power system, the dynamic process such as gas pressure needs to be considered, so dynamic simulation needs to be performed during each power system dispatch cycle, and the simulation time step is selected. In the process of joint simulation, an interactive mode based on event and time double driving is adopted.

[0091] Table 2 Power-natural gas system joint simulation process

[0092] 1) Event-driven In this embodiment, the event types include: a) Initial failure of system elements due to extreme disasters. Including: failure of lines, transformers and generators in the power system; gas supply capacity loss, compressor failure, etc. in the natural gas system. Since natural gas pipelines are generally built underground, it is assumed that extreme disasters will not cause pipeline damage or freezing.

[0093] b) Natural gas system cascading failure. Power system failure may trigger natural gas system cascading failure, such as electrically driven facilities exiting operation. After the power system fails, emergency control measures need to be output according to the power system failure evolution model to determine the position of the load shedding and the load shedding amount, and in this process, the power supply to electrically driven facilities may be cut off. At this time, the power system needs to pass parameter information to the natural gas system, including the gas supply amount required by the gas turbine unit F G (see equation (2-14)) and the running state of the electrically driven compressor in the natural gas system u C and the running state of the electrically driven gas source u GS .In this paper, the variables reflecting the running state of the elements are all 0-1 integer variables, and 1 is taken if the element is normally running, and 0 is taken if it is faulty.

[0094] c) Power system cascading failure. This embodiment focuses on the exit of gas turbine units due to low pressure at the supply node and the disconnection of lines and load shedding due to emergency control measures of the power system. Gas source or compressor failure in the natural gas system can also cause the pressure at the gas turbine unit supply node to decrease, and the pressure at the gas turbine unit supply node needs to be detected after each step in the natural gas system simulation. At this time, the gas network needs to pass parameter information to the power grid, including the running state of the gas turbine unit u G and the power consumption of electrically driven facilities P C and P GS (see equations (2-15)-(2-16)).

[0095] 2) Time-driven If the above event does not occur in the simulation period of the natural gas system, data exchange between the two systems is needed to synchronize information. That is, when the natural gas system completes the calculation of one step, wherein , the power consumption required by the electric drive facility needs to be transmitted to the power system. After the power system performs an AC-OPF-based dispatch once, the gas consumption required by the gas turbine is transmitted to the natural gas system, and the natural gas system updates the state and continues to perform simulation of the next period until the simulation period ends. n This embodiment adopts an AC-OPF-based power system fault evolution model, which is used to simulate the load reduction, generator output adjustment and power system fault evolution process after the element failure of the power system, and output emergency control measures. As shown in , the main steps are as follows:

[0096] Step 1: update the power system state according to the fault, and perform island judgment. For each island, the subsequent operation steps are performed. Figure 4 Step 2: detect whether there is a generator in the island. If there is no generator, all loads in the island lose power supply and store the result, and the next island is executed; if there is a generator, Step 3 is executed.

[0097]

[0098] Step 3: detect whether the island is a single-bus system. If it is a single-bus system, calculate the minimum cut-off load and store the result, and execute the next island; if it is not, execute Step 4.

[0099] Step 4: perform AC-OPF calculation. If it converges, calculate the satisfied load and store the result, and execute the next island; if it does not converge, perform AC-PF calculation and determine the maximum flow limit branch, reduce the load of all nodes within the branch radius R by 5%, and loop Step 4. If the AC-OPF still does not converge after 20 maximum loop times, execute Step 5.

[0100] Step 5: disconnect the most serious flow limit branch. Return to Step 1 to re-judge the island.

[0101] It should be noted that this embodiment focuses on the cascading propagation of faults in a long time scale, and the time scale of power system control operation is generally millisecond level, ignoring the execution time of control operation.

[0102] ​​The process design in this step can accurately capture the evolution process of bidirectional coupling failure, for example: it can accurately simulate the power grid line break → electric-driven compressor shutdown → gas network pressure slowly drops → gas turbine unit exits → power grid further load shedding of power grid-gas network cascading failure, for example, "power grid outage → traffic signal failure → road congestion → power grid emergency repair truck delay → delay in failure recovery of power grid-traffic network interaction. The simulation error of bidirectional coupling failure is reduced to an acceptable range. The phased combination of double drive mechanism solves the simulation breakage problem in the prior art, realizes the full-chain automated simulation from extreme event input to important user influence output, and completes the process of failure triggering, coupling interaction and loss calculation without manual intervention, improves the simulation efficiency, and can completely restore the full-chain time evolution law, providing reliable simulation basis for urban lifeline system resilience evaluation and emergency decision-making.

[0103] S6: output the joint simulation results, including the urban power grid failure evolution process, the function loss rate of each lifeline system and the service availability index of important users.

[0104] The service availability index can be obtained by quantifying the service availability index.

[0105] To verify the availability of the method, the following tests are performed: (1) Test system and parameter settings In this paper, the coupling test system is as follows: power grid: adopt the public IEEE39 node power transmission network system, including generators (G1-G2, G4-G5, G7), and gas turbine units (G3, G6, G8-G10), and lines (Bus1-Bus39); natural gas network: adopt the public Belgium 20-node natural gas system, which includes 6 gas sources (GS1-GS6), and 20 nodes (Node1-Node20), and 2 compressors (GC1, GC2); traffic network: a certain local traffic road network; water supply network: adopt a network including 15 nodes; communication network: adopt a communication network including 200 nodes; rail transit system: a certain city subway station.

[0106] Among them, the line Bus9 of the power grid is coupled with the gas source GS2 of the natural gas network, the line Bus8 of the power grid is coupled with the gas source GS6 of the natural gas network, the node Node6 of the natural gas network is coupled with the gas turbine unit G9 of the power grid, the line Bus25 of the power grid is coupled with the compressor GC2 of the natural gas network, the node Node20 of the natural gas network is coupled with the gas turbine unit G10 of the natural gas network, the line Bus29 of the power grid is coupled with the compressor GC1 of the natural gas network, and the node Node15 of the natural gas network is coupled with the gas turbine unit G8 of the power grid.

[0107] (2) Failure evolution analysis For the above test system, the fault evolution process is analyzed. The fault scenario is set as follows: it is assumed that under the influence of extremely cold and freezing disaster, the gas source GS3 of the natural gas system fails due to freezing at 480 min, the lines Bus5-Bus8 and Bus6-Bus7 of the power system fail due to icing at 500 min, another gas source GS6 also fails due to freezing at 540 min, and the generator G4 of the power system exits operation due to freezing at 560 min.

[0108] Taking the power-natural gas network as an example, the gas source GS2 is set to be in pressure control mode, and the remaining gas sources, gas storage facilities and loads are in flow control mode; the minimum pressure of the gas turbine units G3 and G6 is 5 MPa, and the minimum pressure of the gas turbine unit G9 is 6 MPa; the control mode of the compressor adopts fixed outlet pressure control; the compressor cannot continue to work normally when power is off, and the gas storage facility will affect its natural gas output when power is off. It is assumed that the gas source GS3 is cut off at 3:00 due to extremely cold weather, and the generator G4 fails to start at 6:30 due to weather reasons, and the simulation time is 24 hours.

[0109] In the overall process of fault evolution, after the initial fault of the coupled system caused by extreme weather, if the power system and the natural gas system cannot be well coordinated, it may lead to fault cross-space cascading propagation and aggravate system failure. From the event, the total load shedding of the power system is 2025.53 MW, and one line is removed through the action of the protection device.

[0110] At the beginning of the event, the gas source GS3 of the natural gas system fails due to freezing at 480 min, and the power system and the natural gas system are both normal at this time. Due to the loss of the gas source, the overall pressure level of the natural gas system decreases slightly. At 500 min, the lines Bus5-Bus8 and Bus6-Bus7 of the power system are disconnected due to freezing rain, causing the line Bus9-Bus39 to exceed the limit most seriously, and the power grid lines Bus1, Bus8 and Bus9 cut off the load of 65.11 MW, 365.40 MW and 4.55 MW respectively, totaling 435.06 MW. When the power system performs load shedding operation, the key infrastructure in the natural gas system is not listed as key load. Since the gas source GS2 relies on the power supply of the power system line Bus9, it exits operation due to the emergency control measures of the power system, leading to fault cascading propagation.

[0111] Another gas source GS6 of the natural gas system also failed due to icing at 540 min, causing the pressure level of the system nodes to start to decrease significantly, with the pressure of gas turbine units G3 and G9 changing more obviously. G6 was not affected by the failure of the gas source because it was connected downstream of the compressor GC2. After 143 minutes, the gas turbine unit G9 exited operation because the pressure was lower than the threshold value. At this time, the power system took emergency control measures, and the power grid lines Bus3, Bus4, and Bus8 cut off loads of 111.87 MW, 175 MW, and 54.90 MW, respectively, for a total of 341.77 MW. At 756 minutes, the gas turbine unit G3 also exited operation because the pressure was lower than the threshold value, causing the power system to continue to cut off loads of 571.32 MW, and the compressor GC2 lost power system supply, causing the failure to continue to spread. Because the natural gas system has a slow dynamic characteristic, a failure of a key facility does not cause the pressure to decrease rapidly, and the failure evolution is on a time scale of minutes or even hours, which is the reason why the natural gas dynamic equation needs to be considered in the model.

[0112] After the failure of the compressor GC2, the pressure upstream and downstream of the compressor changed. At the moment of the failure of the compressor, the power consumption of the compressor became 0, and the flow also became 0. After that, the pressure upstream of the compressor rose, and the pressure downstream of the compressor fell. When the pressure upstream of the compressor was equal to the pressure downstream of the compressor, the compressor entered a bypass mode, which was equivalent to a conventional pipeline, and the flow gradually recovered. Because of the failure of the compressor GC2, the pressure of the downstream load decreased, and the pressure of the line Bus20 of the natural gas system fell below the minimum pressure required for the operation of the gas turbine unit G6 at 808 minutes, causing the gas turbine unit G6 to exit operation. The power system had to take emergency control measures again, causing the protection action to remove the overloaded line Bus3-Bus4 and cut off loads of 677.38 MW, and the cascading failure ended.

[0113] During the entire event, this embodiment focused on the output adjustment of the generator unit under the event driving, and the climbing process of the unit was simply considered. At 560 minutes, the generator G4 exited operation. To avoid a larger range of power outages due to load reduction, the output of the remaining units was increased accordingly. Subsequently, three gas turbine units sequentially failed due to excessively low pressure at 683 min, 756 min, and 808 min, causing the failure to intensify.

[0114] In summary, an extreme event can cause the infrastructure of the power or natural gas system to fail sequentially, which can destroy more seriously in actual disasters and then cause the failure to spread across regions and time. In addition, the failure evolution can be accelerated due to the lack of coordination between the power system and the natural gas system, and the test example of this embodiment shows the propagation process. Because of the slow dynamic characteristic of the natural gas system, the failure propagation is relatively slow, and if the above-mentioned characteristic is fully considered, a sufficient failure blocking strategy can be developed to delay or even block the failure propagation and reduce the loss of power outages.

[0115] Similarly, the power-water network, the power-transport network, and the power-communication network are simulated.

[0116] Embodiment 2 Based on embodiment 1, referring to Figure 6 The embodiment provides an event-power grid-life line-user-oriented disaster impact whole-chain joint simulation system, which is used for implementing the event-power grid-life line-user-oriented disaster impact whole-chain joint simulation method in embodiment 1, and the system comprises the following modules. (1) An extreme event fault scenario generation module, which is used for generating an extreme event scenario, analyzing the action mode of the extreme event on a city power grid, and determining initial fault parameters of the power grid.

[0117] (2) A city life line system modeling module, which is used for modeling the city power grid, a natural gas pipe network, a road traffic network, a water supply network, a public communication network and a rail transit system respectively, and obtaining a city life line system dynamics model.

[0118] (3) A coupling element modeling module, which is used for modeling a plurality of coupling elements between the city life line systems, and obtaining a parameter correlation model.

[0119] (4) A resource dependency relationship modeling module, which is used for modeling a resource dependency relationship of an important user to the life line system.

[0120] (5) A city life line simulation module, which is used for performing iterative simulation on the power grid-natural gas pipe network and the power grid-road traffic network with bidirectional coupling characteristics in a first stage, exchanging coupling point information at a fixed step, performing section simulation on the water supply network and the public communication network influenced by the power grid in a single direction in a second stage, calculating a function loss based on fault parameters, realizing interaction of time driving and event driving, and outputting joint simulation results, including a city power grid fault evolution process, a function loss rate of each life line system and an important user service availability index.

[0121] The above merely describes specific embodiments of the present application, but the protection scope of the present application is not limited to this. Any skilled person in the art can easily think of various equivalent modifications or replacements within the technical range disclosed by the present application, and these modifications or replacements should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A joint simulation method for the entire chain of disaster impacts, encompassing events, power grids, lifelines, and users, characterized in that... include: S1: Generate extreme event scenarios, analyze the impact of extreme events on the urban power grid, and determine the initial fault parameters of the power grid; S2: Model the urban power grid, natural gas pipeline network, road traffic network, water supply network, public communication network and rail transit system respectively to obtain the dynamic model of the urban lifeline system; S3: Model multiple coupled components among urban lifeline systems to obtain a parameter correlation model; S4: Model the resource dependencies of critical users on the lifeline system; S5: In the first stage, iterative simulations are performed on the power grid-natural gas pipeline network and the power grid-road traffic network with bidirectional coupling characteristics, and the coupling point information is exchanged at a fixed step size. In the second stage, cross-sectional simulations are performed on the water supply network and public communication network that are affected by the power grid in one direction, and the functional loss is calculated based on the fault parameters to realize the interaction of time-driven and event-driven approaches. S6: Output joint simulation results, including the urban power grid fault evolution process, the functional loss rate of each lifeline system, and the service availability indicators of important users.

2. The disaster impact co-simulation method for the entire chain of events, power grid, lifeline, and users as described in claim 1, is characterized in that... The modeling process of the aforementioned urban lifeline system dynamics model includes: A steady-state model of AC optimal power flow is obtained by modeling the urban power grid; a simplified dynamic model is obtained by modeling the natural gas pipeline network; a hybrid model combining a macroscopic user equilibrium traffic allocation model and a microscopic vehicle following model is obtained by modeling the road traffic network; a steady-state hydraulic model is obtained by modeling the water supply network; a directed weighted graph theory model is obtained by modeling the public communication network; and a dual-subject model of train and passenger is obtained for the rail transit system; thus, a dynamic model of the urban lifeline system is obtained.

3. The disaster impact co-simulation method for the entire chain of events, power grid, lifeline, and users as described in claim 2, is characterized in that... The hybrid model combining the macroscopic user equilibrium traffic assignment model and the microscopic vehicle following model is modeled as follows: Macro-level user equilibrium traffic allocation model: ， ， ， ， ， In the above formula, For road section a Traffic flow For traffic flow aggregation, Represents a cost or time function. Indicates from the source node r To the target node s Total demand Representing a path l Traffic, For indicator functions, when the path contains road segments a The value is 1 if it is true, and 0 otherwise. Indicates from the source node r To the target node s The path, Indicates constraints; Microscopic vehicle car-following model: ， ， ， In the above formula, Indicates time Safe distance speed, Indicates time The speed of the car in front, express Vehicle spacing at any given time For braking reaction time, To keep up with the car's reaction time, Indicates time speed of decision-making Indicates the maximum speed limit of the vehicle. express Current vehicle speed at any given moment This represents the vehicle acceleration relative to the current speed. Indicates the vehicle's actual speed. For time slices, for Expected distance at any time for The current spatial coordinates of the vehicle.

4. The disaster impact co-simulation method for the entire chain of events, power grid, lifeline, and users as described in claim 2, is characterized in that... The constraints of the hybrid model include: cellular traffic flow balance constraints; restrictions that the traffic flow of resource vehicles from a cell to an adjacent cell is no greater than the traffic flow of the cell; restrictions that the total traffic flow of vehicles from a cell to an adjacent cell is no greater than the upper limit of the cell's traffic flow transfer; and restrictions that the traffic flow from an adjacent cell to a cell... The restrictions include: the total traffic flow not exceeding the cell's traffic flow transfer limit; the total inflow into a cell not exceeding the cell's remaining capacity when traffic congestion occurs; and the restrictions on the initial and final traffic flow states in each cell.

5. The disaster impact co-simulation method for the entire chain of events, power grid, lifeline, and users as described in claim 1, is characterized in that... Step S5 includes: S51: Initialize simulation parameters and set simulation duration. Power grid dispatch cycle and natural gas pipeline network simulation step size ,in It is an integer; S52: If the current time Perform spatiotemporal discretization calculation System status at all times Updated to ; S53: In response to the detection of an event, an event-driven interaction is performed, and the power grid and the gas grid exchange the operating status of the gas turbine unit and the power consumption of the electric drive facility. After the power grid performs AC-OPF scheduling, it feeds back the gas consumption of the gas turbine unit. The event includes an initial fault, a gas grid cascading fault, and a power grid cascading fault. S54: If the current time The execution time-driven interaction involves the gas network transmitting the power consumption of the electric drive facilities to the power grid, and the power grid transmitting the gas consumption to the gas network after performing AC-OPF scheduling. S55: After completing the first stage of bidirectional coupling simulation, input the power grid fault parameters into the water supply network and communication network, calculate the water supply loss and the number of communication node failures, output the lifeline system function loss, and complete the second stage of simulation.

6. The disaster impact co-simulation method for the entire chain of events, power grid, lifeline, and users as described in claim 1, is characterized in that... By using functional modeling or principal modeling, we can model the resource dependencies of important users on the lifeline system.

7. The disaster impact co-simulation method for the entire chain of events, power grid, lifeline, and users as described in claim 1, is characterized in that... The key users mentioned include command and control, communications, news media, data and financial centers, water, heat and gas supply, transportation, medical and health care, and important venues.

8. The disaster impact co-simulation method for the entire chain of events, power grid, lifeline, and users as described in claim 1, is characterized in that... The coupling elements include electric compressors, electric gas sources and gas turbines for the power-natural gas network, traffic lights for the power-road transportation network, water pumps for the power-water supply network, base stations / switches for the power-public communication network, and trains for the power-rail transit system.

9. The disaster impact co-simulation method for the entire chain of events, power grid, lifeline, and users as described in claim 1, is characterized in that... The extreme event scenarios mentioned include natural disasters, cyberattacks, physical attacks, major accidents, and complex chain reactions of extreme events.

10. A joint simulation system for the entire chain of disaster impacts, encompassing events, power grids, lifelines, and users, characterized in that... The system is used to implement the event-grid-lifeline-user-disaster impact joint simulation method as described in any one of claims 1-9, the system comprising: The extreme event fault scenario generation module is used to generate extreme event scenarios, analyze the impact of extreme events on the urban power grid, and determine the initial fault parameters of the power grid. The urban lifeline system modeling module is used to model the urban power grid, natural gas pipeline network, road traffic network, water supply network, public communication network and rail transit system respectively, and obtain the dynamic model of the urban lifeline system; The coupling element modeling module is used to model multiple coupling elements between urban lifeline systems to obtain a parameter correlation model; The resource dependency modeling module is used to model the resource dependencies of critical users on the lifeline system; The urban lifeline simulation module is used to perform iterative simulations of the power grid-natural gas pipeline network and power grid-road transportation network with bidirectional coupling characteristics in the first stage, and to exchange coupling point information at a fixed step size. In the second stage, it performs cross-sectional simulations of the water supply network and public communication network that are affected by the power grid in one direction, calculates functional loss based on fault parameters, realizes the interaction of time-driven and event-driven approaches, and outputs joint simulation results, including the urban power grid fault evolution process, functional loss rate of each lifeline system, and service availability indicators for important users.

Images