An urban power distribution network emergency response decision method based on multi-agent V2G response

Abstract

The application relates to the field of power systems, and particularly discloses a city power distribution network emergency response decision-making method based on multi-agent V2G response, which comprises the following steps: a V2G response estimation model based on multi-agent is constructed by using NetLogo; the V2G response estimation model simulates the behavior of multi-agent by integrating a vehicle following model, an improved discrete choice Logit model and a path planning model; a power distribution network operator transmits V2G response to a personalized travel advisor, the personalized travel advisor collects data of electric vehicles and sends a request to the electric vehicles, and the intention of the electric vehicle owners is judged through the vehicle following model and the improved discrete choice Logit model; a recovery model is constructed by using MILP in Matlab; and the recovery model is used for repairing faults in extreme events. The application has the advantages that the recovery efficiency of the power grid after an extreme event is improved, the operation cost is reduced, and effective decision support is provided for power distribution network operators in actual recovery planning.

Description

Technical Field

[0001] This invention relates to the field of power systems, and in particular to an emergency response decision-making method for urban distribution networks based on multi-agent V2G response. Background Technology

[0002] As global warming intensifies, urban areas are facing increasingly frequent extreme weather events, such as torrential rains, floods, and hurricanes. Power shortages or outages caused by these extreme events pose a significant threat to the lives and economic activities of urban residents. Urban power grids, due to the density and interdependence of their infrastructure, are particularly vulnerable to these disasters.

[0003] The resilience of urban power distribution networks is directly related to the overall resilience of cities in the face of extreme events. Improving the disaster resistance of power grids and rapidly restoring power supply has become an urgent problem for the power industry and urban planners. To address these challenges, it is imperative to utilize interdisciplinary technologies to prevent and mitigate the impact of extreme events on power grids, including preventative decision analysis, disaster damage assessment, and optimization of recovery resources. Recent theoretical research has emphasized the importance of improving the disaster resistance of distribution networks for power grid recovery efforts. Researchers have explored the collaboration of various recovery methods, spanning multiple stages and spatiotemporal dimensions, primarily including topology control, power support, and load extraction, such as network reconfiguration, emergency maintenance personnel dispatch, and mobile emergency resource dispatch.

[0004] The coupling between urban power distribution networks and transportation networks is becoming increasingly close. Utilizing transportation networks as a carrier can effectively support power supply and fault repair within the power distribution network. Furthermore, the widespread adoption of electric vehicles (EVs) brings new challenges and opportunities to their interaction with the power grid. EVs not only increase the load demand on the power grid but also provide a new means of power support and recovery through vehicle-to-grid (V2G) technology. V2G technology allows EVs to feed surplus energy back into the grid, providing ancillary services such as demand-side response, load regulation, frequency regulation, and voltage regulation.

[0005] However, current research on how to effectively integrate V2G technology to enhance grid resilience under extreme events remains limited. Existing research often overlooks the potential benefits of electric vehicle battery functionality for V2G response. Therefore, exploring the synergy between V2G services and transportation and distribution networks, and how to effectively integrate the multi-agent characteristics and actions required for electric vehicle aggregation, has significant research value and application prospects for achieving V2G response and improving grid resilience. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide an emergency response decision-making method for urban power distribution networks based on multi-agent V2G response.

[0007] The objective of this invention is achieved through the following technical solution: an emergency response decision-making method for urban power distribution networks based on multi-agent V2G response, comprising:

[0008] A multi-agent V2G response estimation model was constructed using NetLogo;

[0009] The V2G response estimation model simulates the behavior of multiple agents by integrating a vehicle following model, an improved discrete choice Logit model, and a path planning model. These agents include electric vehicles, charging stations, personalized travel advisors, charging station aggregators, and power grid operators.

[0010] The distribution network operator communicates the V2G response to the personalized travel consultant, who collects data on the electric vehicle and makes a request to it. The consultant determines the electric vehicle's intention through a vehicle following model and an improved discrete choice Logit model. The consultant then provides the responding electric vehicle with a route plan to the charging station through a route planning model. The electric vehicle travels to the charging station according to the planned route and begins discharging.

[0011] A recovery model was built using MILP in Matlab to repair faults in extreme events.

[0012] Specifically, the steps for determining the intention of the electric vehicle owner are as follows:

[0013] Step S1: Determine if the electric vehicle responds; if it responds, proceed to step S2.

[0014] Step S2: The personalized travel consultant will sort the shortest paths from the starting point to the charging station in the path list according to the vertex algorithm;

[0015]

[0016] In the formula, For the shortest path, The location of the electric vehicle. For the location of the charging station, step S3, assume that the initial SOC of the electric vehicle follows a normal distribution with mean σ and standard deviation μ, as shown in the following formula:

[0017]

[0018] In the formula, V i and V i cap For the initial battery capacity and total battery capacity of the electric vehicle, Let K be the initial SOC of electric vehicle i. pri and K com This refers to a collection of private and commercial electric vehicles;

[0019] Step S4: Calculate the speed;

[0020] Step S5: Calculate energy consumption:

[0021]

[0022]

[0023] In the formula, γ represents the energy consumption of an electric vehicle while it is in motion. C These are the regression coefficients corresponding to the driving conditions. Let be the instantaneous velocity at time t. Let be the latitude and longitude coordinates of electric vehicle i at time t; T is the total time of the response process, and Δt represents the time difference;

[0024] Step S6: Calculate the final SOC. When the responding electric vehicle arrives at the charging station, the remaining SOC of the responding electric vehicle is the initial SOC minus the energy consumption of the journey from the starting point to the charging station. The specific formula is as follows:

[0025]

[0026] In the formula, In response to the remaining SOC of electric vehicles, For the initial SOC of electric vehicle i, Energy consumption; Step S7: Calculate discharge;

[0027] Step S8: Calculate the response duration;

[0028] Step S9: Determine the direction of the electric vehicle's driver.

[0029] Specifically, in step S1, the remaining SOC of the electric vehicle is calculated to determine whether the electric vehicle has V2G response capability, as shown in the following formula:

[0030]

[0031] In the formula, A qualified SOC level is above 60%. In response to the remaining SOC of electric vehicles.

[0032] Specifically, in step 24, a vehicle following model is used to simulate electric vehicle interaction in real-world driving scenarios. The speed changes of the electric vehicle are to maintain a safe distance from adjacent entities, including maintaining a safe distance from other electric vehicles and red lights. The electric vehicle senses the distance to other entities ahead. and the safe distance d buffA comparison is made to determine whether it is acceleration, deceleration, or stopping; acceleration and deceleration are set to... and Electric vehicle speed Cannot exceed Specifically, in step S7, it is assumed that the discharge process is uniform, and the available power of the responsive electric vehicle should satisfy the following formula:

[0033]

[0034] In the formula, and These are the minimum and maximum discharge power limits for electric vehicles, respectively. When an electric vehicle supplies power to the grid, if the State of Charge (SOC) drops to the reserved value... Setting it to 20% will stop the battery from discharging.

[0035] Specifically, the formula for calculating the response duration in step S8 is as follows:

[0036]

[0037] in, For the response duration, η dis For discharge efficiency, This refers to the unit's discharge power. and This indicates the V2G travel time and the V2G response end time.

[0038] Specifically, in step S9, an improved discrete choice Logit model is used to capture the decision-making process of electric vehicle owners based on their personal attributes. The V2G response intention of electric vehicle owners is transformed into the following expression:

[0039]

[0040] In the formula These represent the benefits, electricity costs, battery degradation costs, and opportunity costs of electric vehicle i participating in V2G response, respectively; π ch It is the electricity price for discharging and charging, η ch It represents charging efficiency, and τ represents the opportunity and cost coefficients. This represents the economic benefit coefficient for car owners. This represents the daily electricity consumption of electric vehicle i. The decision-making process for electric vehicles involves a rough estimate of benefits and costs, therefore the discharge power is used. Replace with estimated schedulable power Specifically, the recovery model aims to reduce the weighted load reduction and V2G compensation costs during post-event recovery, and the objective function is:

[0041]

[0042] In the formula, m f For load weights, C represents the load reduction after power is restored to node f. V2G,t The aggregator will pay compensation to electric vehicle owners;

[0043] The constraints of the recovery model include:

[0044] Active and reactive power balance constraints of network nodes

[0045]

[0046] p b,t and q b,t These represent the active and reactive power injected into node b, respectively. The reactive power of the generator at node b. Let B be the load reactive power, and let B be the set of all nodes. fb,t and q fb,t For line l fb The active and reactive power transmission power, r fb x fb These represent impedance and reactance, respectively. These represent the active power of the generator, V2G, and load at node b, respectively.

[0047] Node, branch voltage amplitude and branch current limit constraints

[0048]

[0049] In the formula, V b,t Let V be the node voltage of node b at time t. f,t Let be the voltage at node f at time t; E is the set of all fault lines. and For two top beam values; I fb,t Let fb be the current at time t;

[0050] Second-order cone constraint

[0051]

[0052] Where |||2 represents the l2 norm,

[0053] Load reduction constraints

[0054]

[0055] In the formula, and For active and passive load reduction, θ is a constant value used for constraints;

[0056] V2G discharge power constraints:

[0057]

[0058] In the formula, V2G discharge power,

[0059] Radial network constraints

[0060]

[0061] In the formula, l fb The flow direction is determined by binary variables and The decision cannot be 1 at the same time. Then the flow direction is positive; if Then the flow direction is negative, if z fb,t =0, then l fb Error. The positive reference direction of the tributary is represented by α and β. When α = 1 and β = 0, the flow direction is from node f to b.

[0062] Specifically, the method further includes constructing an emergency maintenance personnel scheduling model, wherein the constraints for constructing the emergency maintenance personnel scheduling model are:

[0063] Maintenance space constraints:

[0064]

[0065]

[0066] In the formula, C ERC For the purpose of emergency repair, The fault line is a binary variable, and the number of emergency repair personnel is N. ERC d c,u S represents the total distance of the repair route. OD For each emergency repair worker's drive path, if the repair path of emergency repair worker k is from fault line c to u, then its value is 1 or 0, S F S O S D and S E It is a collection of fault lines, start points, end points, and emergency repair personnel, y c,k This is a binary variable indicating whether the faulty line c has been repaired. If the faulty line c has been repaired by emergency repair personnel k, its value will be 1. For the repair task;

[0067] Describe the driving and maintenance routes of the emergency repair personnel:

[0068]

[0069] In the formula, and It represents the starting and ending points of the destination relative to m. This is the node number of a station in the road network. Emergency repair personnel K departs from the station and drives in the order of the fault lines to complete the repair task. And return to the starting point;

[0070] Constraints on repair time:

[0071]

[0072] Emergency repairman K was at Arrival time at fault line c, time taken After repair, the time required exist The time to reach the next fault line u is h, where h is a positive real number.

[0073] The present invention has the following advantages:

[0074] This invention proposes a method for resilient recovery of urban distribution networks that integrates network reconfiguration, emergency maintenance personnel scheduling models, and V2G response. By accurately simulating and predicting V2G response based on a multi-agent model and using mixed-integer linear programming for coordinated optimization, this strategy not only improves the recovery efficiency of the power grid after extreme events but also reduces operating costs, providing effective decision support for distribution network operators in actual recovery planning. Attached Figure Description

[0075] Figure 1 This is a schematic diagram of the recovery method of the present invention; Detailed Implementation

[0076] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention; that is, the described embodiments are merely some embodiments of the invention, and not all embodiments. The components of the embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0077] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0078] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0079] The present invention will be further described below with reference to the accompanying drawings, but the scope of protection of the present invention is not limited to the following description. Figure 1 As shown, an emergency response decision-making method for urban power distribution networks based on multi-agent V2G response includes:

[0080] A multi-agent V2G response estimation model was constructed using NetLogo;

[0081] The V2G response estimation model simulates the behavior of multiple agents by integrating a vehicle following model, an improved discrete choice Logit model, and a path planning model. These agents include electric vehicles, charging stations, personalized travel advisors, charging station aggregators, and power grid operators. Each agent acts as an independent unit with incomplete information and problem-solving capabilities. There is no global control mechanism in the framework; instead, the estimation process is completed through synchronous parallel computation. The V2G response estimation model simulates the individual behavior and collective V2G response effects of electric vehicles at both micro and macro levels, allowing these two levels to coexist. At the micro level, the behavior and decisions of electric vehicles depend on their own attributes and intentions, such as technical parameters like electric vehicle type, battery capacity, initial state of charge (SOC), speed, driving range, and duration, as well as the driver's intentions, such as travel purpose, battery degradation, and income. Macro-level parameters come from the external environment, such as traffic networks, weather, and the impact and mechanisms of extreme disasters.

[0082] The distribution network operator communicates the V2G response to the personalized travel consultant, who collects data on the electric vehicle and makes a request to it. The consultant determines the electric vehicle's intention through a vehicle following model and an improved discrete choice Logit model. The consultant then provides the responding electric vehicle with a route plan to the charging station through a route planning model. The electric vehicle travels to the charging station according to the planned route and begins discharging.

[0083] A recovery model was built using MILP in Matlab to repair faults in extreme events.

[0084] Furthermore, the specific steps for determining the intention of the electric vehicle owner are as follows:

[0085] Step S1: Determine if the electric vehicle responds; the electric vehicle's battery capacity affects its driving range and determines whether the electric vehicle has V2G response capability. CS aggregators require the remaining SOC of the responding electric vehicle. At least 60% of the energy consumed by an electric vehicle during its journey in response to a V2G request, before reaching the nearest CS (Constant Current Edge), is within acceptable limits. This energy consumption should be within acceptable limits for the State of Charge (SOC) level. If the response rate is above 60%, the electric vehicle is eligible to respond, as shown in the following formula:

[0086]

[0087] In the formula, A qualified SOC level is above 60%. In response to the remaining SOC of the electric vehicle; if there is a response, proceed to step S2;

[0088] Step S2: The personalized travel consultant will sort the shortest paths from the starting point to the charging station in the path list according to the vertex algorithm; the shortest paths in the path list... Electric vehicles and target CS The position was calculated using Dijkstra's algorithm:

[0089]

[0090] In the formula, For the shortest path, The location of the electric vehicle. Location of the charging station; j∈SV2G The set representing the target cs;

[0091] Step S3: Assume that the initial SOC of the electric vehicle follows a normal distribution with mean σ and standard deviation μ, as shown in the following formula:

[0092]

[0093]

[0094] In the formula, V i and V i cap For the initial battery capacity and total battery capacity of the electric vehicle, Let K be the initial SOC of electric vehicle i.pri and K com This refers to a collection of private and commercial electric vehicles;

[0095] Step S4: Calculate speed. Simulate electric vehicle interaction in real-world driving scenarios using a vehicle following model. The electric vehicle's speed changes to maintain a safe distance from adjacent entities, including other electric vehicles and red lights. The electric vehicle senses the distance to other entities ahead. and the safe distance d buff A comparison is made to determine whether it is acceleration, deceleration, or stopping; acceleration and deceleration are set to... and Electric vehicle speed Cannot exceed Step S5: Calculate energy consumption:

[0096]

[0097] In the formula, γ represents the energy consumption of an electric vehicle while it is in motion. C These are the regression coefficients corresponding to the driving conditions. Let be the instantaneous velocity at time t. Let be the latitude and longitude coordinates of electric vehicle i at time t, and T be the total time of the response process;

[0098] Step S6: Calculate the final SOC. When the responding electric vehicle arrives at the charging station, the remaining SOC of the responding electric vehicle is the initial SOC minus the energy consumption of the journey from the starting point to the charging station. The specific formula is as follows:

[0099]

[0100] In the formula, In response to the remaining SOC of electric vehicles, For the initial SOC of electric vehicle i, For energy consumption; Step S7, calculate discharge; Assuming the discharge process is uniform, the available power of the responsive electric vehicle should satisfy the following formula:

[0101]

[0102] In the formula, and These are the minimum and maximum discharge power limits for electric vehicles, respectively. When an electric vehicle supplies power to the grid, if the State of Charge (SOC) drops to the reserved value... Setting it to 20% will stop the battery from discharging.

[0103] Step S8: Calculate the response duration; the formula for calculating the response duration is as follows:

[0104]

[0105] in, For the response duration, η dis For discharge efficiency, This refers to the unit's discharge power. and This indicates the V2G travel time and V2G response end time;

[0106] Step S9: Determine the electric vehicle owner's intention; an improved discrete choice Logit model is used to capture the decision-making process of electric vehicle owners based on their personal attributes. The electric vehicle owner's V2G response intention is transformed into the following expression:

[0107]

[0108] In the formula These represent the benefits, electricity costs, battery degradation costs, and opportunity costs of electric vehicle i participating in V2G response, respectively; π ch It is the electricity price for discharging and charging, η ch It represents charging efficiency, and τ represents the opportunity and cost coefficients. This represents the economic benefit coefficient for car owners. α represents the daily electricity consumption of electric vehicle i during its trips. degra To account for the opportunity cost of participating electric vehicles, the electric vehicle decision-making process involves a rough estimate of benefits and costs, thus incorporating discharge power. Replace with estimated schedulable power If EV owners leave their vehicles idle, they will weigh the economic benefits of participating in V2G responses to decide whether to participate.

[0109] Simulation analysis at the macro level

[0110] Personalized travel advisors broadcast V2G response signals to qualified electric vehicles; the actual number of responding electric vehicles... The number of electric vehicles that a charging station's CS (Consumer-to-Grid) capacity can accommodate is limited by the number of charging piles with V2G (Vehicle-to-Grid) functionality. Furthermore, in the initial development phase of V2G response, the main participants are private electric vehicles, accounting for the vast majority of the entire electric vehicle market. In addition, private electric vehicles' daily online interaction with the grid accounts for over 80% of their operating time, enabling them to more effectively receive V2G response signals. Commercial electric vehicles also have significant potential as a response resource.

[0111] Step 1: Establish the traffic network. In the NetLogo platform, the TN is described as a directed graph consisting of nodes and edges representing geographic locations, major roads, and intersections, composed of patches. As patches change along the planned route, electric vehicles travel within the TN from a location patch to a charging station (CS) patch, which can lead to traffic congestion.

[0112] Step 2: Configure V2G response capability. When the responding electric vehicle i arrives at charging station CSj, provide dispatchable power to charging station CSj. The total dispatchable power of electric vehicles is aggregated in CS j at time t and sent to the city distribution network (DN). The CS aggregator pays compensation C to the electric vehicle owners. V2G,t This includes the cost per unit of electricity and the cost of a single response. The specific formula is as follows:

[0113]

[0114] Where, π V2G The compensation price per unit of electricity, π rep This is compensation for a single response.

[0115] Step 3: Set V2G pricing. In emergency situations, electricity market mechanisms can fail. If pricing mechanisms follow normal market conditions, V2G is unlikely to respond to prices, making it difficult to attract support for electric vehicles. Furthermore, distribution network operators sequentially employ recovery methods, including network reconfiguration, emergency maintenance personnel dispatch, and V2G technology. Implementing network reconfiguration and emergency maintenance personnel dispatch provides optimal topology and fault line repair plans. Multi-agent modeling of the V2G response estimates the potential power supply from electric vehicles. The repair aims to minimize weighted load shedding and V2G compensation costs during post-event recovery. The recovery model aims to reduce weighted load shedding and V2G compensation costs during post-event recovery, with the objective function being:

[0116]

[0117] In the formula, m f For load weights, C represents the load reduction after power is restored to node f. V2G,t The aggregator will pay compensation to electric vehicle owners;

[0118] The constraints of the recovery model include:

[0119] Active and reactive power balance constraints of network nodes

[0120]

[0121] p b,t and q b,tThese represent the active and reactive power injected into node b, respectively. The reactive power of the generator at node b. Let B be the load reactive power, and let B be the set of all nodes. fb,t and q fb,t For line l fb The active and reactive power transmission power, r fb x fb These represent impedance and reactance, respectively. The active power of the generator, V2G, and load at node b are respectively; the node and branch voltage amplitude and branch current limit constraints are also specified.

[0122]

[0123] In the formula, V b,t Let V be the node voltage of node b at time t. f,t Let be the voltage at node f at time t; E is the set of all fault lines. and For two top beam values; I fb,t Let fb be the current at time t;

[0124] Second-order cone constraint

[0125]

[0126] Where |||2 represents the l2 norm,

[0127] Load reduction constraints

[0128]

[0129]

[0130] In the formula, and For active and passive load reduction, θ is a constant value used for constraints;

[0131] V2G discharge power constraints:

[0132]

[0133] In the formula, For V2G discharge power, S V2G For discharge power constraint;

[0134] Radial network constraints

[0135]

[0136] In the formula, l fb The flow direction is determined by binary variables and The decision cannot be 1 at the same time. Then the flow direction is positive; if Then the flow direction is negative, if z fb,t =0, then l fb An error occurred. The positive reference direction of the tributary is represented by α and β. When α = 1 and β = 0, the flow direction is from node f to b, E. slack For the constraint on z.

[0137] Furthermore, the method also includes constructing an emergency maintenance personnel scheduling model, wherein the constraints for constructing the emergency maintenance personnel scheduling model are:

[0138] Maintenance space constraints:

[0139]

[0140]

[0141] In the formula, C ERC For the purpose of emergency repair, The fault line is a binary variable, and the number of emergency repair personnel is N. ERC d c,u S represents the total distance of the repair route. OD For each emergency repair worker's drive path, if the repair path of emergency repair worker k is from fault line c to u, then its value is 1 or 0, S F S O S D and S E It is a collection of fault lines, start points, end points, and emergency repair personnel, y c,k This is a binary variable indicating whether the faulty line c has been repaired. If the faulty line c has been repaired by emergency repair personnel k, its value will be 1. For the repair task;

[0142] Describe the driving and maintenance routes of the emergency repair personnel:

[0143]

[0144] In the formula, and It represents the starting and ending points of the destination relative to m. This is the node number of a station in the road network. Emergency repair personnel K departs from the station and drives in the order of the fault lines to complete the repair task. And return to the starting point;

[0145] Constraints on repair time:

[0146]

[0147] Emergency repairman K was at Arrival time at fault line c, time taken After repair, the time required exist The time is to reach the next fault line u, where h is a very large positive real number.

[0148] The above description is merely a preferred embodiment of the present invention and does not constitute any limitation on the present invention. Any person skilled in the art can make many possible variations and modifications to the technical solution of the present invention, or modify it into equivalent embodiments, without departing from the scope of the present invention. Therefore, any modifications, equivalent changes, and alterations made to the above embodiments based on the technology of the present invention without departing from the scope of the present invention are within the protection scope of the present invention.

Claims

1. A decision-making method for emergency response in urban power distribution networks based on multi-agent V2G response, characterized in that: include: A multi-agent V2G response estimation model was constructed using NetLogo; The V2G response estimation model simulates the behavior of multiple agents by integrating a vehicle following model, an improved discrete choice Logit model, and a path planning model. These agents include electric vehicles, charging stations, personalized travel advisors, charging station aggregators, and power grid operators. The distribution network operator communicates the V2G response to the personalized travel consultant, who collects data on the electric vehicle and makes requests to it. The consultant then uses a vehicle following model and an improved discrete choice Logit model to determine the electric vehicle's intention. The path planning model provides the electric vehicle with a route to the charging station, and the electric vehicle travels to the charging station according to the planned route to start discharging. A recovery model is built using MILP in Matlab for fault repair in extreme events. The specific steps for determining the intention of the electric vehicle owner are as follows: Step S1: Determine if the electric vehicle responds; if it responds, proceed to step S2. Step S2: The personalized travel consultant will sort the shortest paths from the starting point to the charging station in the path list according to the vertex algorithm; ； In the formula, For the shortest path, The location of the electric vehicle. Location of the charging station; Step S3: Set the initial SOC of the electric vehicle to follow the mean. and standard deviation The normal distribution is as follows: ； ； In the formula, and For the initial battery capacity and total battery capacity of the electric vehicle, For electric vehicles The initial SOC, and This refers to a collection of private and commercial electric vehicles; Step S4: Calculate the speed; Step S5: Calculate energy consumption: ； ； In the formula, This refers to the energy consumption of an electric vehicle while it is in motion. These are the regression coefficients corresponding to the driving conditions. for Instantaneous speed at a moment For electric vehicles exist Latitude and longitude coordinates of the moment This is the total time taken for the response process; Step S6: Calculate the final SOC. When the responding electric vehicle arrives at the charging station, the remaining SOC of the responding electric vehicle is the initial SOC minus the energy consumption of the journey from the starting point to the charging station. The specific formula is as follows: ； In the formula, In response to the remaining SOC of electric vehicles, For electric vehicles The initial SOC, Energy consumption; Step S7: Calculate the discharge; Step S8: Calculate the response duration; Step S9: Determine the direction of the electric vehicle driver; In step S9, an improved discrete choice Logit model is used to capture the decision-making process of electric vehicle owners based on their personal attributes. The V2G response intention of electric vehicle owners is transformed into the following expression: ； ； ； ； ； ； ； In the formula, , , , They represent electric vehicles. Benefits of participating in V2G response, electricity costs, battery degradation costs, and opportunity costs; Is it the electricity price for discharging or charging? It's about charging efficiency. This represents the opportunity cost coefficient. This represents the economic benefit coefficient for car owners. Indicates electric vehicles Daily electricity consumption for travel To account for the opportunity cost of participating electric vehicles, the electric vehicle decision-making process involves a rough estimate of benefits and costs, thus incorporating discharge power. Replace with estimated schedulable power .

2. The urban distribution network emergency response decision-making method based on multi-agent V2G response as described in claim 1, characterized in that: In step S1, the remaining SOC of the electric vehicle is calculated to determine whether the electric vehicle has V2G response capability, as shown in the following formula: ； In the formula, A qualified SOC level is above 60%. In response to the remaining SOC of electric vehicles.

3. The emergency response decision-making method for urban power distribution networks based on multi-agent V2G response as described in claim 1, characterized in that: In step S4, a vehicle following model is used to simulate electric vehicle interaction in a real-world driving scenario. The electric vehicle's speed changes to maintain a safe distance from adjacent entities, including other electric vehicles and red lights. The electric vehicle senses the distance to other entities ahead. and its safe distance A comparison is made to determine whether it is acceleration, deceleration, or stopping; acceleration and deceleration are set to... and The speed of electric vehicles Cannot exceed .

4. The emergency response decision-making method for urban power distribution networks based on multi-agent V2G response as described in claim 1, characterized in that: In step S7, it is assumed that the discharge process is uniform, and the available power of the responsive electric vehicle should satisfy the following formula: ； ； In the formula, and These are the minimum and maximum discharge power limits for electric vehicles, respectively. When an electric vehicle supplies power to the grid, if the State of Charge (SOC) drops to the reserved value... When set to 20%, the battery will stop discharging.

5. The urban distribution network emergency response decision-making method based on multi-agent V2G response as described in claim 4, characterized in that: The formula for calculating the response duration in step S8 is as follows: ； ； in, For response duration, For discharge efficiency, This refers to the unit's discharge power. and This indicates the V2G travel time and V2G response end time.

6. The emergency response decision-making method for urban power distribution networks based on multi-agent V2G response as described in claim 1, characterized in that: The recovery model aims to reduce the weighted load reduction and V2G compensation costs during post-event recovery, and the objective function is: ； In the formula, For load weights, For nodes Load reduction after power restoration The aggregator will pay compensation to electric vehicle owners; The constraints of the recovery model include: Active and reactive power balance constraints of network nodes ； ； ； ； In the formula, and They are nodes b The active and reactive node injected power, For nodes b The reactive power of the generator is at that location. For the reactive power of the load, The set of all nodes. and For the line The active and reactive power transmission power, , These represent impedance and reactance, respectively. , , They are nodes b Active power of generators, V2G and loads; Node, branch voltage amplitude and branch current limit constraints ； ； ； In the formula, For nodes exist The node voltage at that time, for time The voltage at the node; It is the set of all fault lines; and For two top beam values; for t Time fault line The current; Second-order cone constraint ； in, Represents the l2 norm, Load reduction constraints ； ； ； In the formula, and For active and passive load reduction, This is a constant value used for constraints; V2G discharge power constraints: ； In the formula, This refers to the V2G discharge power. Radial network constraints ； ； ； In the formula, The flow direction is determined by binary variables and The decision cannot be 1 at the same time. If the flow direction is positive, then the flow direction is positive; if If the flow direction is negative, then the flow direction is negative. ,but Error, the positive reference direction of the tributary is used. and It means that when and At that time, the flow direction is from node arrive , To constraint.

7. The emergency response decision-making method for urban power distribution networks based on multi-agent V2G response as described in claim 1, characterized in that: The method further includes constructing an emergency maintenance personnel scheduling model, wherein the constraints for constructing the emergency maintenance personnel scheduling model are: Maintenance space constraints: ； ； ； ； ； ； ； ； ； In the formula, For the purpose of emergency repair, The number of emergency repair personnel is a binary variable representing the fault line. , This represents the total distance of the repair route. For each emergency repair personnel, the drive path, if the emergency repair personnel The repair path is from the fault line arrive If the value is 1 or 0, then its value is either 1 or 0. , , and It is a collection of fault lines, starting points, ending points, and emergency repair personnel. It is a binary variable indicating whether the fault line has been repaired. Emergency repair personnel If repaired, its value will be 1. For the repair task; Describe the driving and maintenance routes of the emergency repair personnel: ； ； ； In the formula, and Is it the destination? The starting point and the ending point, It is the node number of a station in the road network, for emergency repair personnel. Drive from the station and complete the repair tasks following the order of the fault lines. And return to the starting point; Constraints on repair time: ； Emergency repair personnel k exist Arrive at the fault line at any time c ,time consuming After repair, the time required exist Time to reach the next fault line ,in, It is a positive real number.

Images