A cold weather V2G emergency scheduling method fusing multi-dimensional indexes

Abstract

The application discloses a kind of cold weather V2G emergency scheduling method of fusion multidimensional index, belong to electric vehicle and power grid interaction technical field, including the following steps: S1, the state data of power-traffic coupling system is collected;S2, cold weather influence model is constructed;S3, all electric vehicles are screened, and generate to be scheduled charging candidate set and can participate in discharge candidate set;S4, multidimensional evaluation index is constructed;S5, according to state data, cold weather influence model and multidimensional evaluation index, to be scheduled charging candidate set and can participate in discharge candidate set are handled, and V2G emergency charge-discharge scheduling instruction is generated.The present application is introduced through ingenious multi-objective reward function design and economic incentive mechanism, while pursuing system level target, also fully consider the individual interests and service experience of electric vehicle user, greatly improve the social acceptability of scheduling strategy and the promotion feasibility in real world.

Description

Technical Field

[0001] This invention belongs to the field of electric vehicle and power grid interaction technology, specifically involving a V2G emergency dispatch method for cold weather that integrates multi-dimensional indicators. Background Technology

[0002] In cold weather conditions, such as cold waves and blizzards, power and transportation systems are particularly vulnerable. On the one hand, low temperatures can cause electrical equipment (such as transformers and switches) to become brittle, and transmission and distribution lines are prone to icing, leading to widespread distribution network failures and power outages for critical loads such as hospitals, communication base stations, and electric vehicle charging stations. On the other hand, cold weather significantly impacts the performance and usage of electric vehicles: the available capacity of the power battery decreases sharply, while the demand for heating from the vehicle's air conditioning increases dramatically, resulting in a significant reduction in the actual driving range of electric vehicles. This exacerbates "range anxiety" for electric vehicle users, making charging needs more concentrated and urgent.

[0003] Vehicle-to-Grid (V2G) technology utilizes a massive number of electric vehicles as distributed mobile energy storage units, providing a new technological approach to address the aforementioned challenges. By dispatching V2G-enabled electric vehicles to faulty areas or critical loads for emergency discharge, power supply can be quickly restored at a lower cost, ensuring the basic operation of the system. Current research has explored the use of V2G for grid emergency recovery, primarily employing mathematical programming or reinforcement learning methods to formulate dispatch strategies.

[0004] However, existing V2G emergency dispatching solutions have significant limitations when dealing with the specific and complex scenario of extreme weather, greatly reducing the feasibility and effectiveness of their strategies in real-world applications. First, existing methods generally ignore the severe impact of extreme weather on the physical characteristics of electric vehicles. For example, in cold weather, the charging and discharging efficiency of electric vehicle batteries decreases significantly, charging time is greatly extended, and the demand for heating from the vehicle's air conditioning leads to a sharp increase in driving energy consumption. Most dispatching models still use idealized parameters under normal temperature conditions. This disconnect from physical reality makes their dispatching plans (such as the expected discharge capacity and dispatching time) impractical and difficult to execute effectively in real-world environments.

[0005] The existing dispatching system oversimplifies the behavioral response model of electric vehicle users. V2G emergency dispatching relies on the voluntary participation of vehicle owners, whose decisions are influenced by multiple factors such as economic incentives, battery degradation, and personal time costs. Existing technologies often assume that all eligible vehicles will participate in dispatching unconditionally, or use a fixed response rate, lacking a response model that can dynamically predict user participation willingness based on real-time economic incentives. This leads to a significant overestimation of the scale of dispatchable V2G resources.

[0006] Current research largely relies on simulations based on macroscopic or static traffic models, failing to adequately consider the dynamic impact of V2G scheduling behavior on microscopic traffic flow. For example, large-scale scheduling of electric vehicles to the same faulty charging station inevitably alters traffic flow in the surrounding road network, potentially triggering new traffic congestion. This congestion, in turn, increases the travel time and energy consumption of subsequently scheduled vehicles. Existing methods lack a system capable of real-time closed-loop simulation of V2G scheduling and microscopic traffic dynamics, thus failing to assess and mitigate the potential negative traffic effects of scheduling strategies.

[0007] Therefore, there is an urgent need for a V2G emergency power supply system and method specifically designed for cold weather scenarios. This system can comprehensively consider multiple physical constraints in harsh environments and perform online closed-loop control in a high-fidelity power-traffic dynamic coupling environment, thereby formulating an emergency dispatch strategy that is truly feasible and has system-level optimization effects. Summary of the Invention

[0008] To address the above problems, this invention proposes a V2G emergency dispatch method for cold weather that integrates multi-dimensional indicators.

[0009] The technical solution of this invention is: a V2G emergency dispatch method for cold weather that integrates multi-dimensional indicators, comprising the following steps:

[0010] S1. Collect status data of the power-transportation coupled system;

[0011] S2. Construct a model of the impact of cold weather;

[0012] S3. Screen all electric vehicles to generate a candidate set of charging to be scheduled and a candidate set of discharging to participate in the discharge process;

[0013] S4. Construct multi-dimensional evaluation indicators;

[0014] S5. Based on the status data, the cold weather impact model, and multi-dimensional evaluation indicators, process the candidate sets of charging to be scheduled and the candidate sets of discharging that can participate, and generate V2G emergency charging and discharging scheduling instructions.

[0015] Furthermore, in S1, the status data includes the operating status and ambient temperature of charging stations in the power system, as well as the state of charge and location of each electric vehicle in the transportation system.

[0016] Furthermore, in S2, the cold weather impact model includes a temperature-dependent charging model and a cold weather in-transit energy consumption model;

[0017] The expression for the temperature-dependent charging model is:

[0018] ;

[0019] In the formula, t represents the given charging duration, T represents the ambient temperature, S0 represents the initial charging level, S(t) represents the final charging level, β0 represents the first coefficient, β1 represents the second coefficient, β2 represents the third coefficient, and ln(·) represents the logarithmic function.

[0020] Cold Weather On-the-Go Energy Consumption Model E total The expression is:

[0021] ;

[0022] ;

[0023] ;

[0024] ;

[0025] ;

[0026] ;

[0027] In the formula, E cruise E represents the energy required to overcome resistance while moving at a constant speed. aux F represents the energy consumption caused by the auxiliary power of the air conditioner. total η represents the total resistance, s represents the stroke length, and η represents the total resistance. prop F represents propulsion efficiency. air F represents air resistance. roll ρ represents rolling resistance, A represents air density, and C represents the projected area at the front of the vehicle. d The value represents the air drag coefficient, m represents the vehicle mass, g represents the acceleration due to gravity, and C represents the acceleration due to gravity. rr P represents the rolling resistance coefficient. aux denoted by , where represents auxiliary power, and v represents average speed.

[0028] Furthermore, in S3, the electric vehicle's state of charge is detected. If the state of charge is less than the first threshold and the electric vehicle's original planned journey does not pass through a charging station, then the electric vehicle belongs to the candidate set of vehicles to be scheduled for charging; if the state of charge is greater than the second threshold, then the electric vehicle belongs to the candidate set of vehicles that can participate in discharging.

[0029] Furthermore, S4 includes the following sub-steps:

[0030] S41. Generate routing efficiency indicators, queuing balance indicators, emergency power supply cost indicators, and traffic delay indicators;

[0031] S42. Construct a multi-dimensional evaluation index based on routing efficiency index, queuing balance index, emergency power supply cost index, and traffic delay index.

[0032] Furthermore, in S41, the expression for the routing efficiency metric R1 is:

[0033] ;

[0034] In the formula, l j The shortest distance between the location of electric vehicle j when it receives a dispatch and the target charging station is shown. j,min l represents the distance to the nearest charging station from the current location of electric vehicle j. j,max n is the distance to the charging station furthest from the current location of electric vehicle j. ev This indicates the number of electric vehicles currently being dispatched;

[0035] In S41, the expression for the queuing equilibrium index R2 is:

[0036] ;

[0037] ;

[0038] ;

[0039] In the formula, t i Let n represent the equivalent queuing time at charging station i. ch n represents the number of electric vehicles queuing for charging at the corresponding charging station. disch P represents the number of electric vehicles queuing at the corresponding charging station. i This represents the charging and discharging power of the charging station, in μ. wait This represents the average equivalent queuing time for charging stations in the area, and n represents the number of charging stations.

[0040] In S41, the expression for the emergency power supply cost index R3 is:

[0041] ;

[0042] ;

[0043] ;

[0044] In the formula, C g p represents the cost of using an emergency generator to generate electricity. g n represents the equivalent power generation price per kW·h of emergency generators. cs P represents the number of charging stations. i This indicates the charging and discharging power of the charging station, t inter C represents the time interval between each interaction between the agent and the environment. disch C indicates payment for the discharge vehicle. incent For V2G incentive price, p e This indicates the electricity price at this time;

[0045] In S41, the expression for the traffic delay index R4 is:

[0046] ;

[0047] ;

[0048] In the formula, n edge t represents the number of roads adjacent to all charging stations. f This indicates the travel time (t) of this road segment when there is no congestion and the flow is free. BPR The actual travel time of the road segment is represented by Q, the traffic flow is represented by C, the road capacity is represented by α, the first parameter to be calibrated is represented by β, and the second parameter to be calibrated is represented by β.

[0049] In S42, the multidimensional evaluation index R total The expression is:

[0050] ;

[0051] In the formula, ω1 represents the weight coefficient of the routing efficiency index, ω2 represents the weight coefficient of the queuing balance index, ω3 represents the weight coefficient of the emergency power supply cost index, and ω4 represents the weight coefficient of the traffic delay index.

[0052] Furthermore, in S5, based on state data and the cold weather impact model, and using the maximization of multidimensional evaluation indicators as the metric, the intelligent control model generates V2G emergency charging and discharging scheduling instructions from the candidate set of charging to be scheduled and the candidate set of discharging that can participate.

[0053] Furthermore, in S5, the intelligent control model is generated using a multi-agent dual-deep Q-network algorithm.

[0054] The beneficial effects of this invention are:

[0055] (1) The present invention significantly improves the realism and accuracy of the control strategy: The present invention explicitly and quantitatively models the negative impact of cold weather on the energy consumption and charging time of electric vehicles, enabling the scheduling strategy to make decisions based on predictions that are closer to physical reality, thereby avoiding scheduling plan failures caused by model inaccuracies and significantly improving the accuracy and reliability of control.

[0056] (2) The present invention effectively enhances the resilience of critical services of the power grid: In the simulated charging station failure scenario, the present invention can successfully use the electric vehicle fleet as a virtual power plant to provide continuous power support for critical charging stations that have lost external power by intelligent V2G emergency dispatch of a large number of electric vehicles. This greatly alleviates the power supply pressure of the power grid and ensures the continuity of critical charging services.

[0057] (3) This invention realizes deep synergistic optimization of the power and transportation systems: This invention innovatively quantifies the congestion impact of V2G scheduling decisions on the transportation system and uses it as one of the optimization objectives. This enables the control model to actively avoid schemes that may cause traffic congestion when making scheduling decisions, thereby improving the resilience of the power grid and maintaining the operating efficiency of the transportation system, and truly realizing the synergistic gains of the two systems.

[0058] (4) The present invention takes into account the interests of all parties and improves the feasibility of the strategy: Through the ingenious design of multi-objective reward function and the introduction of economic incentive mechanism, the present invention, while pursuing system-level goals (such as power grid resilience and traffic efficiency), also fully considers the individual interests and service experience of electric vehicle users (such as reducing unnecessary routing and queuing time), which greatly improves the social acceptability of the scheduling strategy and its feasibility for promotion in the real world. Attached Figure Description

[0059] Figure 1 A flowchart for a V2G emergency dispatch method for cold weather that integrates multi-dimensional indicators;

[0060] Figure 2 This is a schematic diagram illustrating the dynamic changes in the number of charging and discharging vehicles at charging station 1 and charging station 2 during the fault time in an embodiment of the present invention.

[0061] Figure 3 This invention presents the specific performance of the emergency scheduling strategy and random scheduling strategy based on multi-agent reinforcement learning in four evaluation metrics in this embodiment of the invention. Detailed Implementation

[0062] The embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0063] like Figure 1As shown, this invention provides a V2G emergency dispatch method for cold weather that integrates multi-dimensional indicators, including the following steps:

[0064] S1. Collect status data of the power-transportation coupled system;

[0065] S2. Construct a model to assess the impact of cold weather;

[0066] S3. Screen all electric vehicles to generate a candidate set of charging to be scheduled and a candidate set of discharging to participate in the discharge process;

[0067] S4. Construct multi-dimensional evaluation indicators;

[0068] S5. Based on the status data, the cold weather impact model, and multi-dimensional evaluation indicators, process the candidate sets of charging to be scheduled and the candidate sets of discharging that can participate, and generate V2G emergency charging and discharging scheduling instructions.

[0069] This invention aims to accurately model the unique impact of cold weather on the behavior of electric vehicles and construct a comprehensive evaluation system that includes multiple dimensions such as power grid, transportation and economy. Through intelligent decision-making, it achieves synergistic optimization of these conflicting objectives, so as to effectively cope with the risk of cascading failures under extreme weather conditions and improve the service resilience and overall operational efficiency of urban critical infrastructure.

[0070] In this embodiment of the invention, in S1, the status data includes the operating status and ambient temperature of the charging station in the power system, as well as the charge status and location of each electric vehicle in the transportation system.

[0071] In this embodiment of the invention, in step S2, a series of mathematical models are established to describe the impact of cold weather on the behavior of electric vehicles, including a temperature-dependent electric vehicle charging time model and an in-transit energy consumption model considering onboard air conditioning heating. The cold weather impact model includes a temperature-dependent charging model and a cold weather in-transit energy consumption model;

[0072] Taking into full account the impact of different temperatures on the charging time of electric vehicles at urban fast charging stations, the expression for the temperature-dependent charging model is as follows:

[0073] ;

[0074] In the formula, t represents the given charging duration, T represents the ambient temperature, S0 represents the initial charging level, S(t) represents the final charging level, β0 represents the first coefficient, β1 represents the second coefficient, β2 represents the third coefficient, and ln(·) represents the logarithmic function.

[0075] t is in minutes, T is in degrees Celsius, and the coefficients β0, β1, and β2 are 0.015, 0.00034, and −0.022, respectively.

[0076] During emergency dispatching in cold weather, electric vehicles need to travel from their current location to the target charging station. During this time, the vehicle not only needs to provide energy to the drive system but also consumes a significant amount of electrical energy for in-vehicle heating, leading to a significant decrease in its remaining State of Charge (SOC). Based on the classic power load model and combined with modeling the energy consumption of auxiliary equipment, a simplified energy consumption calculation framework is constructed: the cold weather on-the-go energy consumption model E. total The expression is:

[0077] ;

[0078] ;

[0079] ;

[0080] ;

[0081] ;

[0082] ;

[0083] In the formula, E cruise E represents the energy required to overcome resistance while moving at a constant speed. aux F represents the energy consumption caused by the auxiliary power of the air conditioner. total η represents the total resistance, s represents the stroke length, and η represents the total resistance. prop F represents propulsion efficiency. air F represents air resistance. roll ρ represents rolling resistance, A represents air density, and C represents the projected area at the front of the vehicle. d The value represents the air drag coefficient, m represents the vehicle mass, g represents the acceleration due to gravity, and C represents the acceleration due to gravity. rr P represents the rolling resistance coefficient. aux denoted by , where represents auxiliary power, and v represents average speed.

[0084] In this embodiment of the invention, in S3, the electric vehicle's state of charge is detected. If the state of charge is less than a first threshold and the electric vehicle's original planned route does not pass through a charging station, then the electric vehicle belongs to the candidate set of charging to be scheduled; if the state of charge is greater than a second threshold, then the electric vehicle belongs to the candidate set of discharging that can participate.

[0085] At each decision-making moment, all electric vehicles in the network are automatically traversed. First, the state of charge (SOC) of each vehicle is checked; by default, a SOC below 20% indicates a charging need. If the vehicle's original route does not pass through a charging station, it is added to the list requiring dispatch. Second, to verify energy adequacy, if a vehicle's SOC is greater than 80%, it is assumed to have discharge capability and can enter the discharge candidate set. Subsequently, these candidate vehicles will partially participate in V2G emergency power supply based on their responsiveness to economic incentives. This tiered screening process efficiently filters vehicles with charging and discharging needs and capabilities from a massive number of electric vehicles. By aggregating supply and demand resources in the transportation network in real time, it provides robust and accurate data support for decision-making, thereby ensuring the safe and stable operation of the urban transportation-electricity coupling system during cold weather.

[0086] In this embodiment of the invention, S4 includes the following sub-steps:

[0087] S41. Generate routing efficiency indicators, queuing balance indicators, emergency power supply cost indicators, and traffic delay indicators;

[0088] S42. Construct a multi-dimensional evaluation index based on routing efficiency index, queuing balance index, emergency power supply cost index, and traffic delay index.

[0089] This invention designs a comprehensive reward function for evaluating the merits of scheduling strategies. This function is formally represented by an evaluation index system encompassing multiple dimensions. The index system includes at least the following four dimensions: 1) a routing efficiency index to quantify the additional travel costs incurred due to scheduling, thereby improving user experience and reducing energy consumption; 2) a queuing balance index to quantify the load balance among charging stations, thereby improving the overall utilization efficiency of charging facilities; 3) an emergency power supply cost index to quantify the economic expenditure of ensuring power supply to faulty charging stations; and 4) a traffic delay index to quantify the derivative impact of scheduling decisions on traffic congestion.

[0090] In this embodiment of the invention, in S41, the routing efficiency index is implemented by penalizing the distance between the target charging station of the dispatched vehicle and its current location. The expression for the routing efficiency index R1 is:

[0091] ;

[0092] In the formula, l j The shortest distance between the location of electric vehicle j when it receives a dispatch and the target charging station is shown. j,min l represents the distance to the nearest charging station from the current location of electric vehicle j. j,max n is the distance to the charging station furthest from the current location of electric vehicle j. ev This indicates the number of electric vehicles currently being dispatched;

[0093] l i The solution is obtained using Dijkstra's algorithm.

[0094] In S41, the queuing balance index is achieved by penalizing the deviation between the equivalent queuing time of each charging station and its average queuing time in the network, guiding vehicles to stations with surplus service capacity. The expression for the queuing balance index R2 is:

[0095] ;

[0096] ;

[0097] ;

[0098] In the formula, t i Let n represent the equivalent queuing time at charging station i. ch n represents the number of electric vehicles queuing for charging at the corresponding charging station. disch P represents the number of electric vehicles queuing at the corresponding charging station. i This represents the charging and discharging power of the charging station, in μ. wait This represents the average equivalent queuing time for charging stations in the area, and n represents the number of charging stations.

[0099] In S41, the expression for the emergency power supply cost index R3 is:

[0100] ;

[0101] ;

[0102] ;

[0103] In the formula, C g p represents the cost of using an emergency generator to generate electricity. g n represents the equivalent power generation price per kW·h of emergency generators. cs P represents the number of charging stations. i This indicates the charging and discharging power of the charging station, t inter C represents the time interval between each interaction between the agent and the environment. disch C indicates payment for the discharge vehicle. incent For V2G incentive price, p e This indicates the electricity price at this time;

[0104] The calculation of emergency power supply cost incentives varies depending on the scenario: when the total power provided by dispatched V2G discharging vehicles is sufficient to meet the load demand of the faulty charging station, this cost is simply the sum of the incentive fees paid to all participating electric vehicle users; when the V2G discharging power is insufficient, this cost includes not only the V2G incentive fee but also the high cost of activating a backup mobile emergency generator. This design guides the intelligent control model to prioritize and fully utilize the lower-cost V2G resources.

[0105] In S41, the traffic delay index uses the BPR road impedance function to calculate and penalize the additional travel time delay caused by traffic flow changes due to scheduling; the expression for the traffic delay index R4 is:

[0106] ;

[0107] ;

[0108] In the formula, n edge t represents the number of roads adjacent to all charging stations. f This indicates the travel time (t) of this road segment when there is no congestion and the flow is free. BPR The actual travel time of the road segment is represented by Q, the traffic flow is represented by C, the road capacity is represented by α, the first parameter to be calibrated is represented by β, and the second parameter to be calibrated is represented by β; generally, α is set to 1.5 and β to 4.0.

[0109] In S42, the multidimensional evaluation index R total The expression is:

[0110] ;

[0111] In the formula, ω1 represents the weighting coefficient of the routing efficiency index, ω2 represents the weighting coefficient of the queuing balance index, ω3 represents the weighting coefficient of the emergency power supply cost index, and ω4 represents the weighting coefficient of the traffic delay index. The weighting coefficients are set according to operational preferences.

[0112] In this embodiment of the invention, in step S5, based on the state data and the cold weather impact model, and using the maximization of multidimensional evaluation indicators as the indicator, a V2G emergency charging and discharging scheduling instruction is generated using an intelligent control model from the candidate set of charging to be scheduled and the candidate set of discharging that can participate.

[0113] In this embodiment of the invention, in step S5, the intelligent control model is generated using a multi-agent dual-depth Q-network algorithm.

[0114] The present invention will now be described with reference to specific embodiments.

[0115] On a SUMO-based microscopic traffic simulation platform, an integrated power system model is used. The system collects the real-time status of the entire network in 10-minute decision cycles. The collected data includes: (a) real-time traffic flow at each intersection; (b) spatiotemporal distribution of charging demand for electric vehicles; (c) time-of-use electricity pricing; (d) V2G incentive pricing; (e) real-time ambient temperature; and (f) the SOC distribution of all electric vehicles in the road network. In this embodiment, charging stations 1 and 2 are assumed to completely lose external power supply during the critical period from 6:00 to 18:00 due to an upstream power grid failure.

[0116] Generally, m = 1500 kg is taken, and C is taken as... d =0.6, take A=5m², ρ=1.225kg / m³, take C rr =0.01, take g=9.81m / s², take η prop =0.9, take P aux =2000W.

[0117] like Figure 2 As shown, during the fault period from 6:00 to 18:00, for the faulty charging stations 1 and 2, the number of V2G discharge vehicles dispatched by this invention (dashed line in the upper sub-figure / dashed line in the lower sub-figure) can dynamically and intelligently respond to their charging needs (solid line in the upper sub-figure / solid line in the lower sub-figure), effectively providing power support and successfully supporting the charging load of charging stations 1 and 2 at up to 85.66% and 53.10% respectively.

[0118] like Figure 3 As shown, the method of this invention is compared with a random scheduling strategy. The results show that the method of this invention achieves the best performance in all four dimensions: routing reward R1, queuing reward R2, cost reward R3, and delay reward R4, fully demonstrating the significant technical advantages and great practical application value of this method in V2G emergency charging and discharging control under cold weather conditions.

[0119] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of this invention.

Claims

1. A V2G emergency dispatch method for cold weather that integrates multi-dimensional indicators, characterized in that, Includes the following steps: S1. Collect status data of the power-transportation coupled system; S2. Construct a model to assess the impact of cold weather; S3. Screen all electric vehicles to generate a candidate set of charging to be scheduled and a candidate set of discharging to participate in the discharge process; S4. Construct multi-dimensional evaluation indicators; S5. Based on the status data, the cold weather impact model, and multi-dimensional evaluation indicators, process the candidate sets of charging to be scheduled and the candidate sets of discharging that can participate, and generate V2G emergency charging and discharging scheduling instructions.

2. The cold weather V2G emergency dispatch method integrating multi-dimensional indicators according to claim 1, characterized in that, In S1, the status data includes the operating status and ambient temperature of the charging station in the power system, as well as the charge status and location of each electric vehicle in the transportation system.

3. The cold weather V2G emergency dispatch method integrating multi-dimensional indicators according to claim 1, characterized in that, In S2, the cold weather impact model includes a temperature-dependent charging model and a cold weather in-transit energy consumption model. The expression for the temperature-dependent charging model is: ； In the formula, t represents the given charging duration, T represents the ambient temperature, S0 represents the initial charging level, S(t) represents the final charging level, β0 represents the first coefficient, β1 represents the second coefficient, β2 represents the third coefficient, and ln(·) represents the logarithmic function. The cold weather energy consumption model E total The expression is: ； ； ； ； ； ； In the formula, E cruise E represents the energy required to overcome resistance while moving at a constant speed. aux F represents the energy consumption caused by the auxiliary power of the air conditioner. total η represents the total resistance, s represents the stroke length, and η represents the total resistance. prop F represents propulsion efficiency. air F represents air resistance. roll ρ represents rolling resistance, A represents air density, and C represents the projected area at the front of the vehicle. d The value represents the air drag coefficient, m represents the vehicle mass, g represents the acceleration due to gravity, and C represents the acceleration due to gravity. rr P represents the rolling resistance coefficient. aux denoted by , where represents auxiliary power, and v represents average speed.

4. The cold weather V2G emergency dispatch method integrating multi-dimensional indicators according to claim 1, characterized in that, In step S3, the electric vehicle's state of charge is detected. If the state of charge is less than a first threshold and the electric vehicle's original planned route does not pass through a charging station, the electric vehicle belongs to the candidate set of vehicles to be scheduled for charging. If the state of charge is greater than a second threshold, the electric vehicle belongs to the candidate set of vehicles that can participate in discharging.

5. The cold weather V2G emergency dispatch method integrating multi-dimensional indicators according to claim 1, characterized in that, S4 includes the following sub-steps: S41. Generate routing efficiency indicators, queuing balance indicators, emergency power supply cost indicators, and traffic delay indicators; S42. Construct a multi-dimensional evaluation index based on routing efficiency index, queuing balance index, emergency power supply cost index, and traffic delay index.

6. The cold weather V2G emergency dispatch method integrating multi-dimensional indicators according to claim 5, characterized in that, In S41, the expression for the routing efficiency index R1 is: ； In the formula, l j The shortest distance between the location of electric vehicle j when it receives a dispatch and the target charging station is shown. j,min l represents the distance to the nearest charging station from the current location of electric vehicle j. j,max n is the distance to the charging station furthest from the current location of electric vehicle j. ev This indicates the number of electric vehicles currently being dispatched; In S41, the expression for the queuing balance index R2 is: ； ； ； In the formula, t i Let n represent the equivalent queuing time at charging station i. ch n represents the number of electric vehicles queuing for charging at the corresponding charging station. disch P represents the number of electric vehicles queuing at the corresponding charging station. i This represents the charging and discharging power of the charging station, in μ. wait This represents the average equivalent queuing time for charging stations in the area, and n represents the number of charging stations. In S41, the expression for the emergency power supply cost index R3 is: ； ； ； In the formula, C g p represents the cost of using an emergency generator to generate electricity. g n represents the equivalent power generation price per kW·h of emergency generators. cs P represents the number of charging stations. i This indicates the charging and discharging power of the charging station, t inter C represents the time interval between each interaction between the agent and the environment. disch C indicates payment for the discharge vehicle. incent For V2G incentive price, p e This indicates the electricity price at this time; In S41, the expression for the traffic delay index R4 is: ； ； In the formula, n edge t represents the number of roads adjacent to all charging stations. f This indicates the travel time (t) of this road segment when there is no congestion and the flow is free. BPR The actual travel time of the road segment is represented by Q, the traffic flow is represented by C, the road capacity is represented by α, the first parameter to be calibrated is represented by β, and the second parameter to be calibrated is represented by β. In S42, the multidimensional evaluation index R total The expression is: ； In the formula, ω1 represents the weight coefficient of the routing efficiency index, ω2 represents the weight coefficient of the queuing balance index, ω3 represents the weight coefficient of the emergency power supply cost index, and ω4 represents the weight coefficient of the traffic delay index.

7. The cold weather V2G emergency dispatch method integrating multi-dimensional indicators according to claim 1, characterized in that, In step S5, based on state data and the cold weather impact model, and using the maximization of multidimensional evaluation indicators as the indicator, a smart control model is used to generate V2G emergency charging and discharging scheduling instructions from the candidate set of charging to be scheduled and the candidate set of discharging that can participate.

8. The cold weather V2G emergency dispatch method integrating multi-dimensional indicators according to claim 7, characterized in that, In S5, the intelligent control model is generated using a multi-agent dual-deep Q-network algorithm.

Images