A vehicle-grid interaction guidance simulation method and system

Abstract

The present invention discloses a vehicle-grid interaction guidance simulation method and system. The method includes: constructing a multi-dimensional factor set for vehicle-grid interaction, covering factors on the user side, charging pile side, and road side; constructing a vehicle-grid interaction guidance element set to optimize user behavior through a guidance mechanism, improving the charging and discharging efficiency of electric vehicles and the operation efficiency of the power grid; designing a quantification mechanism for the influence of multi-dimensional factors of vehicle-pile-road, and using a multi-time scale attention mechanism to extract the influence characteristics of each factor; optimizing the user-side response mechanism to obtain an optimal response strategy; based on the optimal response on the user side, selecting the optimal guidance mechanism on the grid side to ensure the stable operation of the power grid; updating the parameters related to the influence of multi-dimensional factors of vehicle-grid interaction, and adjusting the guidance strategy to adapt to the dynamic power grid and user behavior. The present invention can improve the charging and discharging efficiency of electric vehicles, optimize the operation and stability of the power grid, and improve the weighted optimum of the user's power anxiety value, travel efficiency, and income, providing two-way optimization decision support for the power grid and users.

Description

Technical Field

[0001] The present invention relates to the technical fields of smart grid and transportation, and particularly to a vehicle-grid interaction guidance simulation method and system. Background Art

[0002] As a clean energy transportation vehicle, electric vehicle (EV) has received extensive attention. The popularization of electric vehicles not only helps to reduce the dependence on traditional fossil fuels and greenhouse gas emissions, but also brings profound changes to the transportation field. However, the rapid development of electric vehicles also brings new challenges and requirements to the operation and management of the power system, especially the smart grid.

[0003] In the traditional power system, the operation of the power grid mainly depends on the stable balance between power generation and power consumption. The charging demand and quantity change of electric vehicles will have a greater impact on the load distribution of the power grid. At the same time, the unbalanced phenomenon between the energy demand during the charging process of electric vehicles and the power grid load may lead to power grid load fluctuations, affecting the operation stability and security of the power grid. Therefore, how to effectively manage the interaction between electric vehicles and the power grid has become an important issue in the current power grid operation management.

[0004] Vehicle-grid interaction technology (V2G, Vehicle-to-Grid) as an emerging solution allows for bidirectional energy flow between electric vehicles and the power grid. Electric vehicles can charge during the low load period of the power grid and feed electrical energy back to the power grid during the high load period to provide energy storage services. This bidirectional energy exchange not only helps to balance the power grid load, improve the peak shaving ability and stability of the power grid, but also brings economic benefits to electric vehicle users, such as obtaining compensation for participating in demand response projects. In addition, vehicle-grid interaction can also improve the energy utilization efficiency of electric vehicles, reduce the operation cost of the power system, and increase the proportion of renewable energy used.

[0005] However, the effective implementation of vehicle-grid interaction technology faces many challenges, especially in the field of vehicle-grid interaction simulation. Although the current simulation technology can simulate the interaction process between electric vehicles and the power grid, there are still some key problems:

[0006] Limitations of single-factor models: Existing vehicle-grid interaction simulation technologies usually only consider the influence of a single factor on the system, and fail to comprehensively consider the coupling effects of multi-dimensional factors (such as power grid load, electric vehicle location, charging pile status, user behavior, etc.) on vehicle-grid interaction. Traditional methods lack accurate modeling based on multi-time scale attention mechanisms and cannot comprehensively quantify the time-varying influence of these factors at different time scales. Therefore, these simulation methods cannot meet the changing charging demands of electric vehicles, nor can they achieve the overall optimization of the vehicle-grid interaction system, resulting in insufficient accuracy and adaptability of the simulation results.

[0007] Lack of dual - perspective optimization: Traditional vehicle - grid interaction simulation methods mostly focus on a single perspective of the power grid or the user side, while ignoring the conversion of the interaction perspective between users and the power grid. The optimization of the vehicle - grid interaction system is not only about the load balance of the power grid, but also involves the personalized needs and economic benefits of electric vehicle users during the charging process. Therefore, the lack of a multi - perspective optimization process that integrates the user - side response mechanism and the power - grid - side guidance mechanism leads to the inability to achieve collaborative optimization between the power grid and users.

[0008] Absence of deviation between theoretical and actual performance: Current vehicle - grid interaction simulation methods generally ignore the deviation update mechanism based on the difference between the theoretically optimized performance value and the actual optimized performance value. The behaviors of the power grid and users are highly dynamic, and simple theoretical optimization performance calculations cannot fully reflect the complex situations in actual operation. Therefore, the lack of real - time update of mechanism parameters results in a large deviation between the actual performance of the simulation system and user requirements or the power - grid operation status, making it impossible to adjust and adapt in a timely manner. Summary of the Invention

[0009] In view of the deficiencies of the prior art, the present invention provides a vehicle - grid interaction guidance simulation method and system. It should not only comprehensively consider the mutual influences among multi - dimensional factors, but also be able to perform two - way optimization between the power grid and users, and adjust strategies in real - time to adapt to dynamic environmental changes. The present invention can more accurately and comprehensively reflect the interaction relationship between electric vehicles and the power grid, providing a new solution for the overall optimization of the vehicle - grid interaction system.

[0010] To achieve the above - mentioned invention purposes, the technical solutions adopted by the present invention are as follows:

[0011] A vehicle - grid interaction guidance simulation method includes the following steps:

[0012] S1: Construct a multi - dimensional factor set for vehicle - grid interaction, considering user - side factors, charging - pile - side factors, and road - side factors respectively. User - side factors include the departure place, destination, vehicle SoC, range - anxiety value, travel efficiency, and revenue. Charging - pile - side factors include charging - pile power, idle status, and service - fee price. Road - side factors include traffic flow, number of lanes, and lane width.

[0013] S2: Construct a set G of vehicle - grid interaction guidance elements o , where the guidance elements include the unit price of ancillary service revenue, charging and swapping service fees, and demand - response revenue, and guide user behaviors through different guidance mechanisms to optimize the charging and discharging behaviors of electric vehicles and the operation efficiency of the power grid.

[0014] S3: Design a quantification mechanism for the influencing forces of vehicle-pile-road multi-dimensional factors, model the influencing forces of factors on the user side, pile side, and road side using a multi-dimensional function model, and extract the influencing force features through a multi-time scale attention mechanism to obtain the influencing force functions of each factor at different time scales.

[0015] S4: Under the influence of multi-dimensional factors, optimize the response mechanism on the user side according to different guiding mechanisms, obtain the optimal response strategy of users, and achieve the weighted optimum of the electricity anxiety value, travel efficiency, and revenue.

[0016] S5: Based on the optimal response mechanism on the user side, the grid side selects the optimal guiding mechanism to minimize the network loss and the source-load imbalance rate, and ensure the stable operation and economy of the power grid.

[0017] S6: Based on the selected optimal guiding mechanism, update the relevant parameters of the influencing forces of vehicle-grid interaction multi-dimensional factors, calculate the theoretical optimization performance values of the power grid and users under this guiding mechanism, and adjust the guiding strategy to adapt to the dynamic power grid operation conditions and user behaviors.

[0018] Furthermore, in step 3, according to historical data and expert experience, fit the single-factor influencing force functions of factors on the user side, pile side, and road side.

[0019] Based on the single-factor influencing force functions, adopt a multi-time scale attention mechanism to obtain the influencing force functions of factors on the user side, pile side, and road side under the influence of multi-dimensional factors.

[0020] For the influencing forces at each time scale, the calculated attention is processed by weighted averaging to obtain the comprehensive influencing forces of each factor at different time scales.

[0021] Furthermore, the single-factor influencing force functions in step 3 are respectively expressed as:

[0022]

[0023] In the formula, are respectively the single-factor influencing force functions of the i-th user-side factor, the j-th pile-side factor, and the k-th road-side factor. is the parameter matrix of the fitting model of the single-factor influencing force function of the i-th user-side factor, the j-th pile-side factor, and the k-th road-side factor. is the fitting model of the single-factor influencing force function of the i-th user-side factor, the j-th pile-side factor, and the k-th road-side factor.

[0024] The influencing force functions of factors on the user side, pile side, and road side are expressed as:

[0025]

[0026] In the formula, are the influence functions of the i-th user-side factor at the annual, monthly, daily, hourly, and minute levels respectively. V Ye,U , V Mo,U , V Da,U , V Ho,U , V Mi,U are the time factors corresponding to the influence of the user-side factor at the annual, monthly, daily, hourly, and minute levels respectively. are the influence functions of the j-th pile-side factor at the annual, monthly, daily, hourly, and minute levels respectively. V Ye,P , V Mo,P , V Da,P , V Ho,P , V Mi,P are the time factors corresponding to the influence of the pile-side factor at the annual, monthly, daily, hourly, and minute levels respectively. are the influence functions of the k-th roadside factor at the annual, monthly, daily, hourly, and minute levels respectively. V Ye,R , V Mo,R , V Da,R , V Ho,R , V Mi,R are the time factors corresponding to the influence of the roadside factor at the annual, monthly, daily, hourly, and minute levels respectively.

[0027] The comprehensive influence of each factor at different time scales is specifically expressed as:

[0028] Comprehensive influence function of the user-side factor at multiple time scales:

[0029]

[0030] Comprehensive influence function of the pile-side factor at multiple time scales:

[0031]

[0032] Comprehensive influence function of the roadside factor at multiple time scales:

[0033]

[0034] In the formula, represents the multi-time scale attention weight of the user-side factor. represents the multi-time scale attention weight of the pile-side factor. represents the multi-time scale attention weight of the roadside factor. represents the influence of the user-side factor and other factor b (such as the mutual influence between other time scales or different factors) at time point t. Indicates the influence of pile-side factors and other factors b at time point t.

[0035] Indicates the influence of roadside factors and other factors b at time point t.

[0036] Furthermore, the user optimal response mechanism in step 4 is based on the multi-dimensional influence function of user-side factors. By solving the optimization problem, the optimal response mechanism set under different guidance mechanisms is obtained, realizing the weighted optimization of the user's personalized electricity anxiety value, travel efficiency, and revenue.

[0037] The optimal response mechanism, according to the response strategy selected by the user, guides the user to charge when the grid load is low by dynamically adjusting the charging price or demand response incentive, reducing the peak load of the grid and improving the benefits of the grid and users.

[0038] Furthermore, the grid-side optimal guidance mechanism selection in step 5 selects the optimal guidance mechanism by minimizing the grid power loss and the source-load imbalance rate, and adjusts the grid operation strategy according to this mechanism.

[0039] The grid-side optimal guidance mechanism selection is based on the optimal response mechanism set, and the optimal mechanism g o is selected to minimize the grid power loss and the source-load imbalance rate, thereby realizing the optimal operation of the grid.

[0040] Furthermore, the influence function update in step 6 includes inputting the guidance mechanism g * into the scenario generation module, calculating the theoretical optimization performance values of the user side and the grid side and using the theoretical optimization performance values to update the parameters of the multi-dimensional factor influence quantification mechanism of vehicle-pile-road, expressed as:

[0041]

[0042] In the formula, is the gradient function. Indicates the gradient with respect to the deviation sample sets Ω User of the user side, pile side, and roadside, Pile Ω Road respectively.

[0043] The present invention also discloses a vehicle-grid interaction guidance simulation system, including: an environment input module, a simulation optimization module, a vehicle-grid interaction guidance simulation module, and a grid control system interface docking module;

[0044] The environment input module is used to input the state and distribution of electric vehicle users, road topology, and environmental information of charging pile data to construct a vehicle-grid interaction simulation scenario;

[0045] The simulation optimization module calculates the optimal response and optimization performance under different guiding mechanisms based on the data provided by the environment input module.

[0046] The vehicle-grid interaction guiding simulation system recommends the optimal response mechanism to users according to the simulation optimization results, optimizing the users' electricity anxiety value, travel efficiency, and benefits.

[0047] The grid control system interface docking module is used to dock the simulation results with the actual power grid system and update the simulation optimization performance value according to the real power grid data.

[0048] Furthermore, the simulation optimization module includes: a guiding mode selection module, a response mechanism selection module, and a simulation scenario generation module;

[0049] The guiding mode selection module is used to select different guiding mechanisms and input them into the system to guide the decision-making of electric vehicle users;

[0050] The response mechanism selection module is used to select the optimal response mechanism of electric vehicle users under different guiding mechanisms;

[0051] The simulation scenario generation module generates corresponding simulation scenarios according to the environmental data provided by the environment input module.

[0052] Furthermore, the vehicle-grid interaction guiding simulation module includes: a user-station-road multi-dimensional influencing factor set construction sub-module, a vehicle-grid interaction guiding element set construction sub-module, a multi-dimensional vehicle-station-road factor influence quantification sub-module, a user-side response mechanism optimization sub-module, a grid-side optimal guiding mechanism selection sub-module, and an influence quantification mechanism parameter update sub-module;

[0053] The user-station-road multi-dimensional influencing factor set construction sub-module constructs an influence set of multi-dimensional factors of users, charging piles, and roads according to the input data provided by the environment input module.

[0054] The vehicle-grid interaction guiding element set construction sub-module is responsible for constructing the vehicle-grid interaction guiding element set.

[0055] The multi-dimensional vehicle-station-road factor influence quantification sub-module constructs a corresponding influence function based on the influence of multi-dimensional factors such as vehicles, stations, and roads on different time scales.

[0056] The user-side response mechanism optimization sub-module calculates the optimal response mechanism from the perspective of the user side.

[0057] The grid-side optimal guiding mechanism selection sub-module: used to evaluate the response effects of electric vehicle users under different guiding mechanisms, quantify the impacts of these responses on the grid operation, and select the guiding mechanism most suitable for the grid demand

[0058] Influence force quantification mechanism parameter update sub-module: By docking with the interface of the actual power grid control system, calculate the deviation between the theoretical optimization performance value obtained in the simulation and the actual optimization performance value, and update the parameters of the multi-dimensional influence function according to the deviation value.

[0059] Compared with the prior art, the advantages of the present invention are as follows:

[0060] 1. The present invention effectively improves the intelligent level of the vehicle-grid interaction system through the time-varying influence force quantification method of vehicle-pile-road multi-dimensional factors based on multi-time scale attention. This method can dynamically capture and quantify the time-varying influence forces of various factors such as vehicles, piles, and roads at different time scales, enabling the simulation system to adapt to changes in factors such as power grid load, traffic flow, and user behavior in real time, enhancing the adaptability and prediction accuracy of the simulation model to environmental changes. In this way, the vehicle-grid interaction system can provide more accurate optimization results, helping electric vehicle users and the power grid achieve collaborative optimization and optimizing the scheduling and utilization efficiency of power resources.

[0061] 2. The proposed guidance mechanism optimization method of the present invention, from the dual perspectives of users and the power grid, takes into account the personalized needs on the user side and the operation needs on the power grid side, and can effectively coordinate the relationship between the two. Through the comprehensive optimization of the user-side response mechanism and the power grid-side guidance mechanism, electric vehicle users can make optimal decisions in aspects such as travel, charging, and demand response, while the power grid can be most effectively regulated in aspects such as power load and network security. This two-way optimization method not only improves the load balancing ability of the power grid but also provides users with a better power consumption experience and economic benefits.

[0062] 3. By comparing the theoretical optimization performance value with the actual optimization performance value and updating the parameters based on the deviation, the present invention significantly improves the actual adaptability of the simulation system. By continuously adjusting the influence function of vehicle-pile-road multi-dimensional factors, the simulation system can be more closely docked with the actual power grid and traffic operation status, making the simulation results closer to actual operations, ensuring that the system can still provide efficient guidance and optimization solutions in a dynamic power grid environment and under changing user behaviors.

[0063] 4. The vehicle-grid interaction guidance simulation method provided by the present invention can not only help the power grid better manage the load and dispatch resources but also encourage users to charge, travel, or participate in demand response during reasonable periods, thereby optimizing the energy distribution of the power grid, reducing power grid fluctuations and load conflicts. At the same time, users can make decisions based on the simulation results to obtain the optimal economic benefits or charging experience. Through this synergy, the operating efficiency of vehicle-grid interaction has been greatly improved, further promoting the coordinated development of electric vehicles and smart grids.

[0064] 5. The present invention optimizes the guidance mechanism and parameter update method, enhancing the flexibility and dynamic adjustment ability of the vehicle-grid interaction guidance strategy. The simulation system can not only cope with the current power grid and traffic conditions but also adjust according to future change trends, ensuring that the strategy remains effective in various situations. This flexible strategy adjustment mechanism provides guarantees for the long-term stability and adaptability of the vehicle-grid interaction system in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 It is a schematic structural diagram of the vehicle-grid interaction guidance simulation system according to an embodiment of the present invention.

[0066] Figure 2 It is a flowchart of the vehicle-grid interaction guidance simulation method according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0067] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the following further elaborates on the present invention with reference to the drawings and by way of examples.

[0068] The present invention provides a vehicle-grid interaction guidance simulation system, as Figure 1 shown, which includes an environmental input module, a simulation optimization module, and a grid control system interface docking module.

[0069] Environmental input module: In the environmental input module, the simulation user can input multi-dimensional factors of vehicle-pile-road to construct a simulation environment. Input the status and distribution of electric vehicle users according to the actual situation, including the location and travel demand of electric vehicle users, etc., to simulate the behavior of electric vehicle users; the simulation user can input the topological structure of the road, including information such as the connection relationship, number of lanes, and lane width of the road, to simulate the actual road network environment; the simulation user can input data such as the location, power, and idle status of charging piles to provide basic charging facility information for the simulation.

[0070] Simulation optimization module: It includes four parts: a guidance mode selection module, a response mechanism selection module, a simulation scenario generation module, and a vehicle-grid interaction guidance simulation module.

[0071] The guidance mode selection module and the response mechanism selection module are the operation interfaces for simulation users in the vehicle-to-grid interaction guidance simulation process. The simulation optimization module can perform simulations according to the preset guidance mechanism. The simulation user can independently select guidance mechanisms such as coupon issuance, electricity price adjustment, and demand response revenue adjustment. The simulation scenario generation module can configure corresponding simulation scenarios based on the environmental inputs of the simulation user. The simulation optimization module uses the proposed vehicle-to-grid interaction guidance simulation method to calculate the optimal response results of electric vehicle users and the theoretical optimization performance values under each mechanism. Then, the system can recommend the optimal response mechanism under the current guidance mechanism to the user based on the simulation optimization results to help electric vehicle users make decisions, and the user can also make their own choices.

[0072] The vehicle-to-grid interaction guidance simulation module includes a multi-dimensional influence factor set construction sub-module for users, charging stations, and roads, a vehicle-to-grid interaction guidance element set construction sub-module, a multi-dimensional vehicle-charging station-road factor influence quantification sub-module, a user-side response mechanism optimization sub-module, a grid-side optimal guidance mechanism selection sub-module, and an influence quantification mechanism parameter update sub-module.

[0073] Multi-dimensional influence factor set construction sub-module for users, charging stations, and roads: Based on the environmental input module, construct a multi-dimensional influence factor set for vehicle-to-grid interaction users, charging stations, and roads.

[0074] Vehicle-to-grid interaction guidance element set construction sub-module: Construct a vehicle-to-grid interaction guidance element set, including guidance mechanisms such as charging and swapping electricity price subsidies and increasing demand response revenue on the grid side.

[0075] Multi-dimensional vehicle-charging station-road factor influence quantification sub-module: Construct a multi-dimensional function model to describe the single-factor influence distribution of vehicle-charging station-road factors at different time points. Based on the influence means, change trends, and change ranges of each factor at large and small time scales, calculate the attention of each factor at large and small time scales. Use multi-time scale attention for feature extraction to obtain the influence functions of user-side, charging station-side, and road-side factors under multi-dimensional factor influence.

[0076] User-side response mechanism optimization sub-module: Calculate the optimal response mechanism under different guidance mechanisms from the user-side perspective, obtain the theoretical optimization performance value, and use it as the theoretical vehicle-to-grid interaction effect. In different guidance mechanism cases, achieve the weighted optimization of the user-side personalized electricity anxiety value, travel efficiency, and revenue by selecting different response mechanisms.

[0077] Grid-side optimal guidance mechanism selection sub-module: Evaluate the response of users under different guidance mechanisms, quantify the impact of different guidance mechanisms on grid operation, and select the optimal guidance mechanism.

[0078] Influence Force Quantification Mechanism Parameter Update Sub-module: Calculate the deviation based on the actual optimization performance value obtained by the Grid Control System Interface Docking Module to update the parameters of the multi-dimensional influence function.

[0079] Grid Control System Interface Docking Module: The Grid Control System Interface Docking Module can dock with the actual grid control system interface to obtain the actual optimization performance value according to the real-time and true vehicle-grid interaction scenario.

[0080] The present invention proposes a vehicle-grid interaction guidance simulation method, as Figure 2 shown, including the following steps:

[0081] Step 1: Construction of the Vehicle-Grid Interaction User-Station-Path Multi-Dimensional Influence Factor Set

[0082] To more comprehensively understand and predict vehicle-grid interaction behavior, this patent fully considers multi-dimensional factors in vehicle-grid interaction and constructs influence factor sets on the user side, station side, and path side respectively. The user-side factors consider the departure place, destination, vehicle SoC, user's personalized electricity anxiety value, travel efficiency, and revenue, denoted as U i (t), i = 1, 2, 3, 4, 5, 6, where i is the index of the user-side factors. The station-side factors include charging pile power, idle status, and service fee price, denoted as P j (t), j = 1, 2, 3, where j is the index of the station-side factors. The path-side factors include traffic flow, number of lanes, and lane width, denoted as R k (t), k = 1, 2, 3, where k is the index of the path-side factors. Among them, t is the time character sequence, t = Year / Month / Day / Hour / Min / Sec, and during simulation, the user can configure the simulation stage, such as t = 2022 / 3 / 10 / 0 / 0 / 0 - 2023 / 3 / 10 / 23 / 59 / 59.

[0083] Step 2: Construction of the Vehicle-Grid Interaction Guidance Element Set

[0084] Vehicle-grid interaction guidance elements can guide and optimize the charging and discharging behavior of electric vehicles during the vehicle-grid interaction process. Through the combined action of these elements, effective energy and information interaction between the power grid and electric vehicles is achieved, improving the operation efficiency and economy of the power grid. The vehicle-grid interaction guidance elements considered in this patent include the unit price of ancillary service revenue, charging and swapping service fees, and demand response revenue. Based on the vehicle-grid interaction guidance elements, the power grid side adopts O types of guidance mechanisms such as charging and swapping electricity price subsidies and increasing demand response revenue, and the guidance mechanism set G o is denoted as G o = {g1,..., g o ,..., g O}, where g o represents the adoption of the o-th guidance mechanism.

[0085] Step 3: Design of the Quantification Mechanism for the Influence of Multi-Dimensional Vehicle-Pile-Road Factors

[0086] 1) To quantify the influence of factors such as users, charging piles, and roads over time, the present invention first constructs a multi-dimensional function model to describe the distribution of the single-factor influence of vehicle-pile-road factors at different time points. Based on expert experience, historical data is used to fit the influence functions of vehicle-pile-road factors, expressed as

[0087]

[0088] In the formula, are the single-factor influence functions of the i-th user-side factor, the j-th pile-side factor, and the k-th road-side factor, respectively. is the parameter matrix of the fitting model of the single-factor influence function of the i-th user-side factor, the j-th pile-side factor, and the k-th road-side factor, is the fitting model of the single-factor influence function of the i-th user-side factor, the j-th pile-side factor, and the k-th road-side factor.

[0089] 2) The single-factor influence function can describe the changes in the influence of vehicle-pile-road factors and time. To more accurately reflect the influence of each factor in the multi-dimensional influence environment of vehicle-network interaction user-charging pile-road, the present invention solves the mutual influence between factors on this basis, obtains the dependence between factors, and combines multi-time scale attention to obtain the influence function of each factor under the condition of multi-factor coupling, so as to quantify the influence of multi-dimensional vehicle-pile-road factors.

[0090] To accurately measure the relative magnitude of the influence of each factor, based on the influence mean value, change trend, and change range of each factor at different time scales, the attention of each factor at different time scales is calculated. Considering that the attention at a smaller time scale can more effectively analyze the rapid changes in the influence of factors in the short term from data characteristics, which is particularly important for application scenarios that require rapid response such as traffic flow prediction. The attention at a larger time scale can help identify long-term trends and periodic patterns, which is helpful for long-term planning and strategy formulation.

[0091] The present invention solves the multi-time scale attention based on the simulated user requirements, including the attention at the minute (Min) level the attention at the hour (Hour) level the attention at the day (Day) level the attention at the month (Month) level the attention at the year (Year) level Taking the solution of minute-level attention as an example, the solution processes of attention at other levels are similar. Minute-level attention is updated based on minutes (Min), and the time window contains T0 data of the current minute. At the minute-level time scale, the minute-scale attention of user-side, pile-side, and road-side factors is calculated according to the mean value, change trend, and change range of the single-factor influence within a time window. It is expressed as:

[0092]

[0093] In the formula, Min_Attention(·) is the attention mechanism function for the small time scale. is the small time-scale attention of the i-th user-side factor, is the small time-scale attention of the j-th pile-side factor, is the small time-scale attention of the k-th road-side factor. is the set of the mean values of the data of the i-th user-side factor in the current and the previous and next time windows, is the set of the linear fitting slopes of the data of the i-th user-side factor in the current and the previous and next time windows, is the set of the ranges of the data of the i-th user-side factor in the current and the previous and next time windows, is the set of the mean values of the data of the j-th pile-side factor in the current and the previous and next time windows, is the set of the linear fitting slopes of the data of the j-th pile-side factor in the current and the previous and next time windows, is the set of the ranges of the data of the j-th pile-side factor in the current and the previous and next time windows, is the set of the mean values of the data of the k-th road-side factor in the current and the previous and next time windows, is the set of the linear fitting slopes of the data of the k-th road-side factor in the current and the previous and next time windows, is the set of the ranges of the data of the k-th road-side factor in the current and the previous and next time windows. Considering the mean value, slope, and range of the influence function comprehensively can evaluate the attention from multiple aspects such as the average state, change direction and rate over time, and stability and volatility of the single-factor influence.

[0094] Based on the single-factor influence functions of user-side, pile-side, and road-side factors, multi-time-scale attention is used for feature extraction to obtain the influence functions of user-side, pile-side, and road-side factors under the influence of multi-dimensional factors It is expressed as

[0095]

[0096] In the formula, They are the influence functions of the $i$-th user-side factor at the annual, monthly, daily, hourly, and minute levels. $V$ Ye,U 、$V$ Mo,U 、$V$ Da,U 、$V$ Ho,U 、$V$ Mi,U They are the time factors corresponding to the influence of the user-side factors at the annual, monthly, daily, hourly, and minute levels. They are the influence functions of the $j$-th pile-side factor at the annual, monthly, daily, hourly, and minute levels. $V$ Ye,P 、$V$ Mo,P 、$V$ Da,P 、$V$ Ho,P 、$V$ Mi,P They are the time factors corresponding to the influence of the pile-side factors at the annual, monthly, daily, hourly, and minute levels. They are the influence functions of the $k$-th roadside factor at the annual, monthly, daily, hourly, and minute levels. $V$ Ye,R 、$V$ Mo,R 、$V$ Da,R 、$V$ Ho,R 、$V$ Mi,R They are the time factors corresponding to the influence of the roadside factors at the annual, monthly, daily, hourly, and minute levels.

[0097] Taking the minute-level influence function of the $i$-th user-side factor as an example, based on the mutual correlation degrees of other factors with factor $i$, solve which is expressed as

[0098]

[0099] In the formula, is the mutual influence. The higher the mutual influence between factor $i$ and other factors, the lower the independence of this factor, the more it depends on other variables, and the smaller its minute-level influence on the system. The annual, monthly, daily, and hourly influence functions of other user-side factors are solved in the same way.

[0100] Similarly, taking the minute-level influence function of the $j$-th pile-side factor as an example, based on the mutual correlation degrees of other factors with factor $j$, solve which is expressed as

[0101]

[0102] The annual, monthly, daily, and hourly influence functions of other pile-side factors are solved in the same way.

[0103] Taking the minute-level influence function of the $k$-th roadside factor as an example, based on the mutual correlation degrees of other factors with factor $j$, solve which is expressed as

[0104]

[0105] The annual, monthly, daily, and hourly influence functions of other roadside factors are solved in the same way as above.

[0106] Step 4: Construction of the optimal response mechanism on the user side.

[0107] In specific implementation, a guidance mechanism can be adopted to encourage users to take specific behaviors. For example, by adjusting the charging price or providing demand response incentives, users are guided to charge when the grid load is low, thereby reducing the peak load of the grid. Under each guidance mechanism, users choose the most suitable response strategy according to their own preferences and actual situations. The present invention calculates the optimal response mechanism under different guidance mechanisms from the perspective of the user side based on the multi-dimensional factor influence of vehicle-pile-road, obtains the theoretical optimization performance value, and takes it as the theoretical vehicle-grid interaction effect. In different guidance mechanism cases, by selecting different response mechanisms, the weighted optimization of the personalized power anxiety value, travel efficiency, and revenue on the user side is achieved. The optimization problems under different guidance mechanisms are constructed as follows from the user side

[0108]

[0109] In the formula, are the influence functions under the multi-dimensional factor influence of the user's personalized power anxiety value, travel efficiency, and revenue respectively, is the v-th response mechanism selected by the user under the condition of adopting the guidance mechanism g o . According to the optimization problem constructed on the user side, the set of optimal response mechanisms under different guidance mechanisms can be obtained Among them, is the optimal response mechanism selected by the user under the condition of adopting the guidance mechanism g o . τ is the simulation time configured by the simulation user.

[0110] Step 5: Selection of the optimal guidance mechanism on the grid side

[0111] The goal on the grid side is to minimize the network loss and the source-load imbalance rate to ensure the stable operation and economy of the grid. Based on the construction of the optimal response mechanism on the user side, the grid side needs to evaluate the response of users under different guidance mechanisms, quantify the impact of different guidance mechanisms on grid operation, and select the optimal guidance mechanism. The grid side can dynamically adjust the guidance strategy to adapt to the changing grid operation conditions and user behaviors.

[0112] The present invention constructs an optimization problem for selecting the optimal guidance mechanism on the grid side. Substituting the optimal response mechanisms selected by users under different guidance mechanisms, the grid side selects different guidance mechanisms to minimize the network loss and the source-load imbalance rate on the grid side. The optimization problem is constructed as

[0113]

[0114] In the formula, Loss(τ) and Ratio(τ) are the network loss and the source-load imbalance rate, which is the weighted sum of the network loss and the source-load imbalance rate when adopting the optimal response mechanism. Based on the optimal response mechanism on the grid side, different guiding mechanisms are introduced, and the guiding mechanism g that minimizes the network loss and the source-load imbalance rate on the grid side is selected * .

[0115] Step 6: Update the parameters of the multi-dimensional factor influence quantification mechanism of vehicle-pile-road

[0116] 1) Adopt the guiding mechanism g * , and input the guiding mechanism g * into the scenario generation module to generate a simulation scenario. Under the simulation scenario, the theoretical optimization performance values of the user side and the grid side are calculated according to formula (5) At the same time, the guiding mechanism g * is used to generate a real scenario in the real-world network through the grid control system, and then it is docked with the simulation system to calculate the actual optimization performance values of the user side and the grid side and judge the deviation situation;

[0117] 2) Calculate the deviation between the theoretical optimization performance value and the actual optimization performance value, and include the samples with a relative deviation greater than the threshold in the set Ω to be updated User , Ω Pile and Ω Road . When , the theoretical and actual deviations of the corresponding user's personalized electricity anxiety value, travel efficiency, and revenue are put into the set Ω User for updating the function parameters on the user side, where |·| is the absolute value function. When , the theoretical and actual deviations of the corresponding travel efficiency and revenue are put into the set Ω Pile for updating the function parameters on the pile side. When , the theoretical and actual deviations of the corresponding network loss and source-load imbalance rate are put into the set Ω Load for updating the function parameters on the road side.

[0118] 3) Based on the deviation samples in the set to be updated, update the parameters of the multi-dimensional factor influence quantification mechanism of vehicle-pile-road, which is expressed as:

[0119]

[0120] In the formula, is a gradient function.

[0121] The method according to the present invention described above can be implemented in hardware, firmware, or be implemented as software or computer code that can be stored in a recording medium (such as a CD ROM, RAM, floppy disk, hard disk, or magneto-optical disk), or be implemented as computer code originally stored in a remote recording medium or a non-transitory machine-readable medium and to be downloaded through a network and stored in a local recording medium, so that the method described herein can be stored in such software processing on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as an ASIC or FPGA). It will be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component (e.g., RAM, ROM, flash memory, etc.) that can store or receive software or computer code, and when the software or computer code is accessed and executed by the computer, the processor, or the hardware, it implements a vehicle-to-grid interaction guidance simulation method described herein. In addition, when a general-purpose computer accesses the code for implementing the processing shown herein, the execution of the code converts the general-purpose computer into a dedicated computer for executing the processing shown herein.

[0122] Those of ordinary skill in the art will realize that the embodiments described herein are for helping the reader understand the implementation methods of the present invention, and it should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those of ordinary skill in the art can make various other specific deformations and combinations without departing from the essence of the present invention based on the technical revelations disclosed in the present invention, and these deformations and combinations are still within the protection scope of the present invention.

Claims

1. A vehicle-grid interaction guidance simulation method, characterized in that: It includes the following steps: S1: Construct a multi-dimensional factor set for vehicle-grid interaction, considering user-side factors, charging pile-side factors, and road-side factors respectively. User-side factors include departure location, destination, vehicle SoC, battery anxiety value, travel efficiency, and revenue. Charging pile-side factors include charging pile power, idle status, and service fee price. Road-side factors include traffic flow, number of lanes, and lane width; S2: Construct the set G of vehicle-grid interaction guiding elements o , where the guiding elements include the unit price of auxiliary service revenue, charging and swapping service fees, and demand response revenue, and the user behavior is guided through different guiding mechanisms to optimize the charging and discharging behavior of electric vehicles and the grid operation efficiency; S3: Design a quantification mechanism for the influence of vehicle-pile-road multi-dimensional factors. Use a multi-dimensional function model to model the influence of user-side, pile-side, and road-side factors, and extract influence features through a multi-time scale attention mechanism to obtain the influence functions of each factor at different time scales; Specifically: according to historical data and expert experience, the single-factor influence functions of user-side factors, pile-side factors, and road-side factors are fitted Based on the single-factor influence function, a multi-time-scale attention mechanism is adopted to obtain the influence functions of user-side, pile-side, and road-side factors under the influence of multi-dimensional factors For the influence at each time scale, the calculated attention is processed by weighted average to obtain the comprehensive influence of each factor at different time scales; S4: Under the influence of multi-dimensional factors, optimize the response mechanism on the user side according to different guiding mechanisms to obtain the optimal response strategy of the user, and achieve the weighted optimum of battery anxiety value, travel efficiency, and revenue; S5: Based on the optimal response mechanism on the user side, the grid side selects the optimal guiding mechanism to minimize the network loss and source-load imbalance rate, and ensure the stable operation and economy of the grid; S6: Based on the selected optimal guiding mechanism, update the relevant parameters of the influence of vehicle-grid interaction multi-dimensional factors, calculate the theoretical optimization performance values of the grid and users under this guiding mechanism, and adjust the guiding strategy to adapt to the dynamic grid operation conditions and user behaviors.

2. The vehicle-grid interaction guidance simulation method according to claim 1, wherein: The single-factor influence functions in step 3 are respectively expressed as: In the formula, are the single-factor influence functions of the i-th user-side factor, the j-th pile-side factor, and the k-th road-side factor, respectively; is the parameter matrix of the fitting model of the single-factor influence function of the i-th user-side factor, the j-th pile-side factor, and the k-th road-side factor, is the fitting model of the single-factor influence function of the i-th user-side factor, the j-th pile-side factor, and the k-th road-side factor; Influence functions of user side, pile side, and road side factors Expressed as: Wherein, are the influence functions of the i-th user-side factor at the annual, monthly, daily, hourly, and minute levels respectively; V Ye,U , V Mo,U , V Da,U , V Ho,U , V Mi,U are the time factors corresponding to the influence of the user-side factor at the annual, monthly, daily, hourly, and minute levels respectively; are the influence functions of the j-th pile-side factor at the annual, monthly, daily, hourly, and minute levels respectively; V Ye,P , V Mo,P , V Da,P , V Ho,P , V Mi,P are the time factors corresponding to the influence of the pile-side factor at the annual, monthly, daily, hourly, and minute levels respectively; are the influence functions of the k-th roadside factor at the annual, monthly, daily, hourly, and minute levels respectively; V Ye,R , V Mo,R , V Da,R , V Ho,R , V Mi,R are the time factors corresponding to the influence of the roadside factor at the annual, monthly, daily, hourly, and minute levels respectively; The comprehensive influence of each factor at different time scales is specifically expressed as: The comprehensive influence function of user-side factors at multi-time scales: The comprehensive influence function of pile-side factors at multi-time scales: The comprehensive influence function of road-side factors at multi-time scales: Wherein, represents the multi-time-scale attention weight of user-side factors; represents the multi-time-scale attention weight of pile-side factors; represents the multi-time-scale attention weight of road-side factors; represents the influence of user-side factors and other factor b at time point t; represents the influence of pile-side factors and other factor b at time point t; represents the influence of road-side factors and other factor b at time point t.

3. The vehicle-grid interaction guidance simulation method according to claim 1, characterized in that: The user optimal response mechanism in step 4 is based on the multi-dimensional influence function of user-side factors. Through solving optimization problems, a set of optimal response mechanisms under different guiding mechanisms is obtained to achieve the weighted optimum of user personalized battery anxiety value, travel efficiency, and revenue; The optimal response mechanism guides users to charge when the grid load is low by dynamically adjusting the charging price or demand response incentive according to the response strategy selected by the user, reducing the peak load of the grid, and improving the benefits of the grid and users.

4. A vehicle-network interaction guidance simulation method according to claim 1, characterized in that: The optimal guiding mechanism selection on the grid side in step 5 selects the optimal guiding mechanism by minimizing the grid network loss and source-load imbalance rate, and adjusts the operation strategy of the grid according to this mechanism; The selection of the optimal guidance mechanism on the grid side is based on the set of optimal response mechanisms, and the optimal mechanism g is selected o to minimize the grid power loss and the source-load imbalance rate of the power grid, thereby achieving the optimal operation of the power grid.

5. A vehicle-grid interaction guidance simulation method according to claim 1, characterized in that: The influence function update in step 6 includes according to the guidance mechanism g * Input to the scenario generation module, and calculate the theoretical optimization performance values on the user side and the grid side And use the theoretical optimization performance values to update the parameters of the vehicle-pile-road multi-dimensional factor influence quantification mechanism, expressed as: In the formula, is the gradient function; represents the gradient with respect to the user side, pile side, and road side deviation sample sets Ω User , Ω Pile , Ω Road .

6. A vehicle-grid interaction guidance simulation system, characterized in that: This system is used to implement the vehicle-grid interaction guiding simulation method described in any one of claims 1 to 5. The system includes: an environment input module, a simulation optimization module, a vehicle-grid interaction guiding simulation module, and a grid control system interface docking module; The environment input module is used to input the status and distribution of electric vehicle users, road topology, and environmental information of charging pile data to construct a vehicle-grid interaction simulation scenario; The simulation optimization module calculates the optimal response and optimization performance under different guiding mechanisms based on the data provided by the environment input module; The vehicle-grid interaction guidance simulation system recommends the optimal response mechanism to users according to the simulation optimization results, and optimizes the users' electricity anxiety value, travel efficiency and benefits. The grid control system interface docking module is used to dock the simulation results with the actual power grid system and update the simulation optimization performance value according to the real power grid data.

7. An interactive vehicle-grid guiding simulation system according to claim 6, characterized in that: The simulation optimization module includes: a guidance mode selection module, a response mechanism selection module, and a simulation scenario generation module. The guidance mode selection module is used to select different guidance mechanisms and input them into the system to guide the decision-making of electric vehicle users. The response mechanism selection module is used to select the optimal response mechanism of electric vehicle users under different guidance mechanisms. The simulation scenario generation module generates corresponding simulation scenarios according to the environmental data provided by the environmental input module.

8. An interactive vehicle-grid guiding simulation system according to claim 6, characterized in that: The vehicle-grid interaction guidance simulation module includes: a user-station-road multi-dimensional influence factor set construction sub-module, a vehicle-grid interaction guidance element set construction sub-module, a multi-dimensional vehicle-station-road factor influence quantification sub-module, a user-side response mechanism optimization sub-module, a grid-side optimal guidance mechanism selection sub-module, and an influence quantification mechanism parameter update sub-module. The user-station-road multi-dimensional influence factor set construction sub-module constructs the influence set of multi-dimensional factors of users, charging piles and roads according to the input data provided by the environmental input module. The vehicle-grid interaction guidance element set construction sub-module is responsible for constructing the vehicle-grid interaction guidance element set. The multi-dimensional vehicle-station-road factor influence quantification sub-module constructs the corresponding influence function based on the influence of multi-dimensional factors of vehicles, stations and roads on different time scales. The user-side response mechanism optimization sub-module calculates the optimal response mechanism from the perspective of the user side. The grid-side optimal guidance mechanism selection sub-module: used to evaluate the response effects of electric vehicle users under different guidance mechanisms, quantify the impacts of these responses on the operation of the power grid, and select the guidance mechanism most suitable for the grid demand. The influence quantification mechanism parameter update sub-module: calculates the deviation between the theoretical optimization performance value obtained in the simulation and the actual optimization performance value by docking with the actual power grid control system interface, and updates the parameters of the multi-dimensional influence function according to the deviation value.

Images