A method for generating charging and discharging scheduling strategies for electric vehicles

Abstract

The present disclosure relates to a method for generating a charging and discharging scheduling strategy for electric vehicles. The method for generating a charging and discharging scheduling strategy for electric vehicles includes: obtaining travel data of an electric vehicle cluster and attribute information of a user cluster corresponding to the electric vehicle cluster; dividing the electric vehicle cluster into at least one intelligent entity according to the attribute information, wherein at least one intelligent entity is determined by classifying electric vehicles with corresponding attributes in the electric vehicle cluster; analyzing and calculating the target travel data of the intelligent entity to determine the charging and discharging information of each electric vehicle in the intelligent entity; obtaining price information of the electricity market during the period corresponding to the target travel data; and generating a charging and discharging scheduling strategy for the intelligent entity with the satisfaction of the user corresponding to the intelligent entity as the target according to the price information and the charging and discharging information. The method provided by the present disclosure generates a simple and effective large-scale charging and discharging scheduling strategy for electric vehicle clusters that can comprehensively consider multiple interest groups.

Description

Technical Field

[0001] The present disclosure relates to the technical field of electric vehicle charging and discharging control, and in particular to a method for generating a charging and discharging scheduling strategy for an electric vehicle. Background Art

[0002] With the continuous iteration and update of electric vehicle products, the number of electric vehicles in use continues to rise. When large-scale electric vehicles are connected to the power grid, there is an impact on the stability of the power system. Therefore, solving the charging and discharging scheduling strategy for large-scale electric vehicle clusters has gradually become a research focus.

[0003] At present, the large-scale electric vehicle charging and discharging scheduling strategy involves more vehicles involved in the scheduling, more variables to be solved, and more difficulty in solving the problem, which makes it more difficult to solve the charging and discharging scheduling strategy and makes it impossible to achieve efficient and reasonable charging and discharging scheduling of large-scale electric vehicles. Therefore, there is an urgent need to provide a simple and effective large-scale electric vehicle charging and discharging scheduling strategy generation method that can comprehensively consider multiple interest groups. Summary of the invention

[0004] In order to solve the above technical problems, the present disclosure provides a method for generating a charging and discharging scheduling strategy for an electric vehicle.

[0005] In a first aspect, an embodiment of the present disclosure provides a method for generating a charging and discharging scheduling strategy for an electric vehicle, comprising:

[0006] Obtaining travel data of the electric vehicle cluster and attribute information of the user cluster corresponding to the electric vehicle cluster;

[0007] Dividing the electric vehicle cluster into at least one intelligent agent according to the attribute information, wherein the at least one intelligent agent is determined by classifying the electric vehicles with corresponding attributes in the electric vehicle cluster;

[0008] Analyze and calculate the target travel data of the intelligent agent to determine the charging and discharging information of each electric vehicle in the intelligent agent;

[0009] Obtain price information of the electricity market during the period corresponding to the target travel data;

[0010] Based on the price information and the charging and discharging information, a charging and discharging scheduling strategy for the intelligent agent is generated with the satisfaction of the user corresponding to the intelligent agent as the goal.

[0011] In a second aspect, the embodiment of the present disclosure provides a device for generating a charging and discharging scheduling strategy for an electric vehicle, comprising:

[0012] A first acquisition unit, used to acquire travel data of an electric vehicle cluster and attribute information of a user cluster corresponding to the electric vehicle cluster;

[0013] A classification unit, used to classify the electric vehicle cluster into at least one intelligent agent according to the attribute information, wherein the at least one intelligent agent is determined by classifying the electric vehicles with corresponding attributes in the electric vehicle cluster;

[0014] A computing unit, used to analyze and calculate the target travel data of the intelligent agent, and determine the charging and discharging information of each electric vehicle in the intelligent agent;

[0015] The second acquisition unit is used to acquire price information of the electricity market during the period corresponding to the target travel data;

[0016] The strategy generation unit is used to generate a charging and discharging scheduling strategy for the intelligent agent with the satisfaction of the user corresponding to the intelligent agent as the target according to the price information and the charging and discharging information.

[0017] In a third aspect, an embodiment of the present disclosure provides an electronic device, including:

[0018] Memory;

[0019] Processor; and

[0020] Computer programs;

[0021] The computer program is stored in the memory and is configured to be executed by the processor to implement the method for generating the electric vehicle charging and discharging scheduling strategy as described above.

[0022] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method for generating a charging and discharging scheduling strategy for an electric vehicle as described above are implemented.

[0023] The embodiment of the present disclosure provides a method for generating a charging and discharging scheduling strategy for electric vehicles, including: obtaining travel data of an electric vehicle cluster and attribute information of a user cluster corresponding to the electric vehicle cluster; dividing the electric vehicle cluster into at least one intelligent entity according to the attribute information, wherein at least one intelligent entity is determined by classifying electric vehicles with corresponding attributes in the electric vehicle cluster; analyzing and calculating the target travel data of the intelligent entity to determine the charging and discharging information of each electric vehicle in the intelligent entity; obtaining price information of the electricity market during the period corresponding to the target travel data; and generating a charging and discharging scheduling strategy for the intelligent entity with the satisfaction of the user corresponding to the intelligent entity as the target according to the price information and the charging and discharging information. The method provided by the present disclosure generates a simple and effective large-scale charging and discharging scheduling strategy for electric vehicle clusters that can comprehensively consider multiple interest groups. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0026] Figure 1 A flow chart of a method for generating a charging and discharging scheduling strategy for an electric vehicle provided in an embodiment of the present disclosure;

[0027] Figure 2 A schematic diagram of the structure of a structurally complex intelligent agent provided in an embodiment of the present disclosure;

[0028] Figure 3 A schematic diagram of a charge and discharge behavior simulation process provided by an embodiment of the present disclosure;

[0029] Figure 4 A schematic diagram of the structure of a device for generating a charging and discharging scheduling strategy for an electric vehicle provided in an embodiment of the present disclosure;

[0030] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0031] In order to more clearly understand the above-mentioned objectives, features and advantages of the present disclosure, the scheme of the present disclosure will be further described below. It should be noted that the embodiments of the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0032] In the following description, many specific details are set forth to facilitate a full understanding of the present disclosure, but the present disclosure may also be implemented in other ways different from those described herein; it is obvious that the embodiments in the specification are only part of the embodiments of the present disclosure, rather than all of the embodiments.

[0033] At present, the simulation methods for electric vehicle travel behavior can be divided into two categories: data-driven modeling and model-based simulation modeling. The data-driven modeling methods mainly include: simulating vehicle charging behavior by collecting historical charging information records of public or private charging facilities, and simulating vehicle driving trajectories through traffic information data provided by road traffic departments. This data-driven method can accurately describe the charging behavior and travel trajectory of vehicles, but it also faces the problems of difficulty in collecting relevant data and complex training of large-scale vehicle clusters. The model-based simulation modeling methods mainly include: modeling probability models and random algorithms based on population trajectory data such as vehicle travel records and global positioning system data, and then simulating vehicle travel trajectories and charging behaviors. Common probability models include normal distribution, stable distribution, and Burr distribution; commonly used random algorithms include Monte Carlo, Markov chain, and kernel density estimation. This model-driven modeling method can describe the overall travel behavior and charging rules of vehicles, and has high computational efficiency, but it also loses the individual differences and population-related characteristics of vehicle users due to its over-generalization. The optimization methods for charging and discharging strategies for electric vehicles can be divided into two categories according to different optimization objectives: single-objective optimization and multi-objective optimization. The single-objective optimization method mainly considers the maximization of vehicle user benefits, such as the shortest vehicle charging time, the minimum charging loss, the minimum charging cost, etc. The multi-objective optimization method mainly considers the combination of multiple objectives such as vehicle user cost, vehicle battery loss, and grid load fluctuation, but lacks the optimization goal of considering the large-scale vehicle cluster charging behavior to absorb the new energy power generation, and does not fully explore the advantages of large-scale vehicle cluster charging and discharging. In short, the current research on vehicle travel simulation lacks an effective method for large-scale vehicle cluster travel simulation that takes into account computing performance, retains population characteristics, and comprehensively considers multiple objectives.

[0034] In summary, the traditional vehicle travel simulation method has certain limitations. First, its simulation accuracy is relatively low. This is mainly because the use of data-driven methods requires a stable and reliable data source, and the simulation accuracy is largely limited by the data quality; the use of model-driven methods is too general, and the simulation accuracy is limited by the typicality and accuracy of the probability distribution model, ignoring the differences in vehicle travel behavior between individuals. In addition, the traditional vehicle travel simulation lacks effective simplification methods for large-scale vehicle clusters and consideration of the demographic attributes of vehicle users. People with different demographic attributes have different work and life patterns, which will lead to differences in vehicle travel behavior among different groups of people. However, the traditional vehicle travel simulation method lacks simplification for large-scale vehicle clusters, relies only on unified model modeling, lacks the characterization and utilization of the demographic attributes of vehicle users, and cannot reflect the diversity of individuals, thus affecting the accuracy of the simulation. The optimization objectives of the traditional vehicle charging and discharging strategy also have certain limitations, lacking consideration of comprehensive objectives, especially the new energy consumption objectives. While the charging and discharging of large-scale vehicle clusters poses greater challenges to the power grid system, it also has its corresponding advantages. The discharge of vehicle clusters has more potential for peak shaving and valley filling for power grid loads, and the charging of vehicle clusters also has more potential for absorbing new energy power generation. The existing traditional methods, in particular, lack optimization considerations for the new energy side and do not fully utilize the charging and discharging advantages of vehicle clusters.

[0035] To address these issues, we proposed a travel simulation method and a charging and discharging optimization strategy generation method for large-scale vehicle clusters. The demographic attributes of vehicle users are used as one of the references for travel simulation and strategy optimization. This can not only effectively classify and simplify large-scale vehicle clusters, but also accurately model the travel behaviors of vehicle groups of different population types on this basis, thereby improving the travel simulation accuracy of vehicle clusters and the computational efficiency of charging and discharging optimization strategies. Comprehensive consideration of vehicle charging and discharging optimization strategies on the user side, grid side, and new energy side is of great significance for tapping the demand-side response potential, ensuring the safe and stable operation of the power grid, promoting the consumption of new energy, and improving energy utilization efficiency.

[0036] Figure 1 A flow chart of a method for generating a charging and discharging scheduling strategy for an electric vehicle provided in an embodiment of the present disclosure, specifically comprising the following steps: Figure 1 The following steps are shown:

[0037] S101. Acquire travel data of an electric vehicle cluster and attribute information of a user cluster corresponding to the electric vehicle cluster.

[0038] Among them, attribute information refers to the portrait information of the user driving the electric vehicle.

[0039] It is understandable that the travel data of the electric vehicle cluster and the attribute information of the user cluster corresponding to the electric vehicle cluster are obtained. The electric vehicle cluster is a large-scale vehicle cluster. The electric vehicle cluster includes multiple electric vehicles. Each electric vehicle has travel data, and the travel data includes travel time, travel distance, charging and discharging behavior, etc. The travel time further includes the travel start time, travel duration and travel end time, etc. Each electric vehicle also has a corresponding user, such as the owner or the user driving the electric vehicle. There are multiple users corresponding to multiple electric vehicles. Multiple users can be understood as the user cluster corresponding to the electric vehicle cluster. Each electric vehicle has attribute information corresponding to the user. The attribute information can be understood as the portrait information of the user driving the electric vehicle. The attribute information includes basic user information and user behavior preferences. The basic information includes the user's age, gender, occupation, etc., which helps to understand the characteristics of the user group. The behavior preference records the user's preferences in choosing charging facilities, travel methods, etc., which are used to customize personalized services.

[0040] S102: Divide the electric vehicle cluster into at least one intelligent entity according to the attribute information.

[0041] Among them, at least one intelligent agent is used to classify and determine the electric vehicles with corresponding attributes in the electric vehicle cluster.

[0042] It is understandable that, based on the above S101, the attribute information of the user cluster is obtained, and the electric vehicle cluster is divided into at least one intelligent entity. Specifically, the division can be carried out using demographic attributes such as gender, age and / or behavioral preferences. The travel behaviors such as travel distance and travel time of electric vehicles corresponding to different population types are different, and this difference will affect the user cluster's choice of electric vehicle charging and discharging preferences. Based on this, all electric vehicles with the same demographic attribute label in the electric vehicle cluster are classified as one intelligent entity, which can form multiple intelligent entities with similar characteristics and needs.

[0043] It is understandable that the travel preferences of electric vehicle users will affect the time for the vehicle to charge and discharge, and the power supply mode of electric vehicles will greatly affect the charging and discharging behavior of electric vehicles. The unified probability model used in the prior art ignores the differences in people's travel habits and is not practical; and the existing research on the power supply mode of electric vehicles is not enough. Most of them only consider the use of charging piles for electric vehicles, while ignoring the possibility of other power supply modes, making the vehicle charging and discharging strategy simplistic. This application proposes to consider the population attributes to model the charging and discharging of electric vehicles with multiple agents. Among them, an agent is an autonomous computer system that can make decisions in a dynamic environment to achieve specific goals. It has four key characteristics: autonomy, sociality, responsiveness and initiative. In a multi-agent system (MAS), multiple agents make decisions and interact together and influence each other. These agents can perceive the environment, learn autonomously and adapt to environmental changes, making the entire system highly flexible and scalable in a dynamic environment. Applying the MAS concept to the electric vehicle charging and discharging system, using electric vehicles as agents, each electric vehicle can choose its own charging and discharging action mode and action time, and vehicles can also communicate with each other to maximize the overall benefits of the vehicle. When facing a large-scale electric vehicle cluster, if each electric vehicle is regarded as an intelligent agent, it will face multiple problems such as increased computational complexity, increased communication overhead between intelligent agents, difficulty in coordination and exploration of intelligent agents, multiple local optimal solutions in the system and difficulty in final convergence. In this regard, the present application adopts a classification method of population attributes to subdivide the electric vehicle cluster. Users with the same population attribute labels often have similar car usage habits and behavior patterns. Therefore, all electric vehicles with the same population attribute labels in the electric vehicle cluster are classified as an intelligent agent. According to the population attribute classification and proportional distribution represented by different intelligent agents in the vehicle cluster, a corresponding trust weight is assigned to each intelligent agent to achieve communication coordination between intelligent agents and maximize the overall interests of the electric vehicle cluster. Finally, the vehicle decision of each population type can be used to represent the decision of all vehicles of that type.

[0044] Optionally, the above-mentioned dividing the electric vehicle cluster into at least one intelligent agent according to the attribute information can be specifically implemented through the following steps:

[0045] According to the gender and age in the attribute information, the target users with the same gender and in the same age range in the user cluster corresponding to the electric vehicle cluster are classified into one category; the target electric vehicles driven by the target users classified into one category are determined in the electric vehicle cluster, and the target electric vehicles are divided into one intelligent entity.

[0046] It can be understood that according to the gender and age in the attribute information, the target users with the same gender and in the same age range in the user cluster corresponding to the electric vehicle cluster are classified into one category, and the same category has the same demographic attribute label. Subsequently, the target electric vehicles corresponding to the target users with the same demographic attribute label in the electric vehicle cluster are classified into one category of intelligent entities. Various demographic attribute labels are shown in Table 1. In Table 1, males with an age range of 25-34 are classified into one category, recorded as the first attribute label, and all electric vehicles corresponding to all users with the first attribute label are regarded as the first intelligent entity. Correspondingly, males with an age range of 35-54 are classified into one category, recorded as the second attribute label, and all electric vehicles corresponding to all users with the second attribute label are regarded as the second intelligent entity. By analogy, 6 categories of intelligent entities can be determined.

[0047] Table 1:

[0048]

[0049] For example, see Figure 2 , Figure 2 A schematic diagram of a structured composite intelligent agent provided in an embodiment of the present disclosure can be used to construct an intelligent agent. The structure of the intelligent agent is as follows: Figure 2 As shown in the figure, it includes eight parts: communication module, information processing module, decision and control module, knowledge base and database, self-learning module, negotiation module, execution module and reflection. This structured intelligent agent can not only respond quickly based on real-time data and environmental changes, but also use historical data and rules in the knowledge base to conduct in-depth analysis and learning through intelligent algorithms to optimize the charging and discharging actions of the vehicle. Overall, it can effectively improve the performance and efficiency of MAS in the vehicle charging and discharging environment and achieve more accurate and efficient management.

[0050] S103: Analyze and calculate the target travel data of the intelligent body to determine the charging and discharging information of each electric vehicle in the intelligent body.

[0051] It can be understood that, based on the above S102, for each intelligent agent, the travel data of all target electric vehicles belonging to each intelligent agent is determined as the target travel data of each intelligent agent. After the target travel data is determined, the target data is analyzed and calculated to simulate the travel behavior of different population attribute groups, and the charging and discharging information of each target electric vehicle in the intelligent agent is determined.

[0052] Optionally, the above analysis and calculation of the target travel data of the intelligent agent to determine the charging and discharging information of each electric vehicle in the intelligent agent can be specifically implemented through the following steps:

[0053] Probabilistic modeling is performed on the travel variables in the target travel data to generate a travel probability model; the travel behavior of each electric vehicle in the intelligent agent is simulated based on the travel probability model, and the travel behavior is at least one of charging behavior, discharging behavior, battery replacement behavior and other behaviors; the travel behavior of each electric vehicle is analyzed and calculated to determine the charging and discharging information of each electric vehicle.

[0054] It can be understood that probability distribution models with different characteristics and simulation focuses are used to perform probability modeling on travel variables in group travel data with different demographic attribute labels, and a travel probability model with demographic attribute labels is generated. The probability distribution models include normal distribution, lognormal distribution, half-normal distribution, stable distribution, Boolean distribution, exponential distribution, gamma distribution, Weibull distribution, inverse Gaussian distribution, log-logistic distribution, logistic distribution, t-location scale distribution, extreme value distribution, generalized extreme value distribution, Birnbaum-Saunders distribution, Nakagami distribution, Rice distribution and Rayleigh distribution, etc. At least one of the above probability distribution models can be used to fit the group travel characteristics with different demographic attribute labels, and the model with the best fitting effect is used as the travel probability model. Specifically, the accuracy of model fitting is evaluated by the evaluation index determination coefficient and the corrected determination coefficient. The evaluation index determination coefficient measures the degree of model fitting data, and the corrected determination coefficient corrects the number of model parameters, eliminating the influence of different numbers of parameters, so that the model fitting evaluation of different numbers of parameters is more fair. The calculation formula of the determination coefficient is shown in formula (1) and formula (2).

[0055]

[0056] In the formula, y i Represents the i-th sample value of the data set y to be fitted. There are n samples in the data set to be fitted. The data set to be fitted refers to the travel data set of all electric vehicles included in the agent, and each sample refers to the travel data of each electric vehicle; represents the sample mean; Represents y i The fitted value is the fitted value output by the probability distribution model; p refers to the number of parameters of the probability distribution model used; R 2 It represents the determination coefficient of the evaluation index; represents the corrected determination coefficient.

[0057] It can be understood that the travel variables are probabilistically modeled through the above-mentioned probability distribution model. After the travel probability model is determined, the travel probability model is used as input, and the Monte Carlo method is used to simulate the travel behavior of user groups with different demographic attribute labels. Then, based on the travel behavior of the user groups, analysis and calculation are performed to obtain charging and discharging information such as the time when electric vehicles participate in charging and discharging actions and the battery state of charge.

[0058] The charge and discharge information includes a historical state reflecting the historical battery charge state, a historical time point corresponding to the historical state, and a current behavior reflecting the current charge and discharge behavior, wherein the current time point corresponding to the current behavior is after the historical time point.

[0059] It can be understood that the charging and discharging information includes the historical status and historical behavior at the historical time point and the current behavior at the current time point. The current behavior occurs after the historical behavior, that is, the current time point is after the historical time point. Among them, the current behavior refers to the specific charging and discharging behavior, such as charging behavior, discharging behavior, battery replacement behavior and other behaviors, and other behaviors can be understood as no behavior, that is, the vehicle has no charging and discharging actions.

[0060] For example, see Figure 3 , Figure 3 A schematic diagram of a charging and discharging behavior simulation process provided for an embodiment of the present disclosure includes: 1) based on the attribute information of the user cluster corresponding to the electric vehicle cluster to be modeled, the population activity travel data corresponding to the user cluster is screened out from the travel data of the electric vehicle cluster, for example, data of the type of charging and discharging activities is screened out; 2) based on the population attribute classification, travel data with different population attribute labels are screened out from the population activity travel data; 3) using the screened data as input, refining the travel start time, travel duration, travel distance and travel end time of the user cluster, and establishing a travel probability model with population attribute labels; 4) using the travel probability model as input, using the Monte Carlo method to simulate the travel behavior of the crowd, and extracting the travel variables to obtain the travel trajectory of the user for the activity throughout the day under the population attribute label; 5) analyzing and calculating based on the user's travel behavior, obtaining charging and discharging information such as the time point and initial charge state of the electric vehicle charging and discharging behavior.

[0061] It is understandable that the power supply modes for electric vehicles are mainly divided into three types: charging, discharging and battery replacement. Different power supply modes have their own advantages and disadvantages in different scenarios. Through reasonable combination and utilization, full coverage and optimization of electric vehicle power supply can be achieved. Therefore, according to the power supply mode of the vehicle, it can be concluded that the travel behavior of the intelligent agent (a certain type of electric vehicle) includes charging, discharging, battery replacement and other behaviors (no behavior).

[0062] S104. Obtain price information of the electricity market during the period corresponding to the target travel data.

[0063] Among them, the price information includes the current price at the current time point and the historical prices at the historical time points. The current price includes the first price for charging electric vehicles, the second price for discharging electric vehicles and the third price for replacing electric vehicle batteries.

[0064] It can be understood that, based on the above S103, the time period corresponding to the target travel data is determined, and the price information of the electricity duration within the time period is obtained. The price information includes the current price at the current time point and the historical price at the historical time point. The current time point and the historical time point are within the time period. The current price refers to the electricity market price of the electric vehicle under different power supply modes. The current price includes the first price when the electric vehicle is charging, the second price when the electric vehicle is discharging, and the third price when the electric vehicle replaces the battery. The first price refers to the charging price, the second price refers to the discharge subsidy price, and the third price refers to the battery replacement price.

[0065] S105. Generate a charging and discharging scheduling strategy for the intelligent agent based on the price information and the charging and discharging information, with the satisfaction of the user corresponding to the intelligent agent as the target.

[0066] It can be understood that on the basis of the above S104, the two key dimensions of the user's economic expenditure and battery power are comprehensively considered, and a reward function with user satisfaction as the goal is established, the price information and charging and discharging information are input into the reward function, and the charging and discharging scheduling strategy of the electric vehicle is formulated according to the specific function value of the reward function. The larger the function value, the higher the user satisfaction. For example, at the current time point, the function value of the electric vehicle's charging behavior is the largest, then it can be recommended that the electric vehicle be charged, and the power grid can also perform corresponding processing.

[0067] Optionally, the above-mentioned generation of a charging and discharging scheduling strategy for an intelligent agent with the satisfaction of a user corresponding to the intelligent agent as a target based on the price information and the charging and discharging information can be specifically implemented through the following steps:

[0068] According to the historical state and historical time points, the current state of the agent under the current behavior is determined, and the current state refers to the battery charge state of the agent at the current time point; according to the current price at the current time point in the price information, the income and expenditure information generated after the agent performs the current behavior at the current time point is calculated; according to the income and expenditure information and the current state, a charging and discharging scheduling strategy for the agent is generated with the satisfaction of the user corresponding to the agent as the goal.

[0069] Understandably, the best indicator to describe the power of electric vehicles is the battery state of charge (SOC), which directly reflects the storage of vehicle energy. According to the historical state (historical state of charge) and historical time points, the current state (current state of charge) of the electric vehicle at the current time point is calculated. The historical time point can be understood as the time point before the current time point. The charging and discharging price of the electricity market is directly related to the economic efficiency of the vehicle charging and discharging process. The electricity price changes with the conditions of the electricity market. On this basis, according to the current price at the current time point in the price information, the income and expenditure information generated by the electric vehicle after performing the current behavior at the current time point is calculated, that is, the cost of charging and discharging behavior is calculated. Subsequently, based on the cost and current state, a reward function with user satisfaction as the target is constructed to guide the intelligent agent to learn the optimal strategy, and generate the charging and discharging scheduling strategy based on the reward function value. Specifically, the reward function is shown in formula (3) and formula (4).

[0070] R(t)={r t 1 ,r t 2 ,r t 3 ,...,r t i ,...,r t n} Formula (3)

[0071]

[0072] In the formula, R(t) represents the function value set of n agents obtained by dividing the electric vehicle cluster at time t, represents the function value of agent i at time t, Indicates the battery charge state (current state), the current time point is moment t, Indicates that agent i performs an action at time t Economic expenditures after It represents the charging and discharging behavior (current behavior) of agent i at time t, β1 and β2 represent the coefficients of battery state of charge and economic expenditure. The coefficient has different values ​​in different time periods and can be set according to user needs.

[0073] Optionally, the above-mentioned determination of the current state of the agent under the current behavior based on the historical state and the historical time point can be specifically implemented through the following steps:

[0074] When the current behavior is not a battery replacement behavior, determine the target increment of the battery state of charge for the current behavior of the intelligent agent at the current time point, and calculate the current state of the intelligent agent under the current behavior based on the target increment and the historical state; or, when the current behavior is a battery replacement behavior, determine the preset state of charge as the current state.

[0075] Understandably, it is determined whether the current behavior is a battery replacement behavior, that is, whether the agent performs a battery replacement action at the current time point. If the current behavior is a battery replacement behavior, the preset state of charge is determined as the battery state of charge (current state) at the current time point. The preset state of charge can be set according to user needs. Preferably, the preset state of charge is 1. If the current behavior is not a battery replacement behavior, that is, the agent may perform charging, discharging, or no action at the current time point, the target increment of the battery state of charge for the agent to perform the current behavior at the current time point is calculated, and the sum of the target increment and the historical state is calculated to obtain the current state of the agent under the current behavior.

[0076] Optionally, the above determination of the target increment of the battery state of charge for the current behavior of the agent at the current time point can be specifically implemented by the following steps:

[0077] When the current behavior is charging behavior, the charging power for executing the charging behavior is determined according to the charging pile that provides charging service for the intelligent body, and the target increment of the battery state of charge of the charging behavior is calculated according to the battery capacity, charging power and the first duration of the continuous charging behavior from the current time point of the intelligent body; or, when the current behavior is discharging behavior, the discharge power for executing the discharging behavior is determined according to the charging pile, and the target increment of the battery state of charge of the discharging behavior is calculated according to the battery capacity, discharge power and the second duration of the continuous discharging behavior from the current time point; or, when the current behavior is other behavior, the preset value is determined as the target increment of the battery state of charge of other behaviors.

[0078] It is understandable that when the current behavior is not a battery replacement action, it is further determined whether the current behavior is a charging behavior, that is, whether the intelligent agent performs a charging action at the current time. If the current behavior is a charging behavior, the charging power is determined according to the charging pile that provides charging services for the intelligent agent. The charging systems of most existing charging stations are discrete, so a discrete charging system can be used to perform charging operations, that is, the charging power and discharging power of electric vehicles can be determined according to the charging pile that provides corresponding services, and the increment of the battery state of charge generated after the charging or discharging action is a discrete value; after determining the charging power, the first duration of the continuous charging behavior starting from the current time point is counted, and the ratio of the product of the first duration and the charging power to the battery capacity is calculated to obtain the increment of the battery charge when the intelligent agent performs the charging action at the current time, wherein the charging power has a positive sign, that is, an identifier with a positive sign is added to the increment to obtain the target increment. If the current behavior is a discharge behavior, the discharge power is determined according to the charging pile, the second duration of the continuous discharge behavior starting from the current time point is counted, and the ratio of the product of the second duration and the discharge power to the battery capacity is calculated to obtain the increment of the battery charge when the intelligent agent performs the discharge action at the current time, wherein the discharge power has a negative sign, that is, an identifier that adds a negative sign to the increment to obtain the target increment; when the current behavior is other than charging, discharging and battery replacement, the preset value is directly determined as the target increment, preferably, the preset value can be 0. Under different charging and discharging behaviors, the formula for calculating the target increment is shown in formula (5).

[0079]

[0080] In the formula, represents the target increment, p ch Indicates charging power, with a positive sign, unit kW, p dc Indicates the discharge power, with a negative sign, in kW, T indicates the duration of the charge / discharge action, in h, E cap Indicates the battery capacity in kWh. Indicates the current behavior. 0 indicates charging behavior. 1 indicates discharge behavior, A value of 3 means no action.

[0081] Optionally, the current state of the agent under the current behavior is calculated based on the target increment and the historical state, which can be achieved through the following steps:

[0082] If the historical time point is the initial time point when the intelligent agent initially participates in scheduling, the current state is calculated based on the target increment and the initial state of the intelligent agent at the initial time point, wherein the historical time point is after the initial time point, and the initial state refers to the remaining charge state of the intelligent agent; or, if the historical time point is not the initial time point, the current state is calculated based on the target increment and the historical state.

[0083] It is understandable that the corresponding battery state of charge update rules are different according to the current behavior of the agent at the current time point. When it is determined that the current behavior is not a battery replacement action, it is determined whether the historical time point is the initial time point when the agent initially participates in scheduling. Initial parameter scheduling refers to scheduling at a pre-set time point throughout the day. For example, the agent initially participates in scheduling twice a day. The initial time point is the time point after work in the morning and after get off work in the evening. If the historical time point is the initial time point, the sum of the initial state and the target increment of the agent at the initial time point is calculated to obtain the current state. The initial state can be understood as the percentage of the remaining power of the electric vehicle; if the historical time point is not the initial time point, the sum of the historical state and the target increment is calculated to obtain the current state. Specifically, the calculation formula of the battery state of charge of the agent at the current time point is shown in formula (6) and formula (7).

[0084]

[0085] In the formula, represents the charge state of agent i at time t, that is, the current state, Represents agent i at t ini The state of charge at that time is the initial state. That is, the remaining power percentage, t ini represents the initial time point, which can be further divided into t ini_1 and t ini_2 , t ini_1 and t ini_2 represents the time points at which agent i initially participates in scheduling twice in a day, It represents the increment of the battery state of charge caused by the behavior of agent i at time t, that is, the target increment. Time t represents the current time point, and the increment has positive and negative signs due to different behaviors, that is, charging is positive, discharging is negative, and no movement is zero. If the battery is replaced, it is directly It represents the state of charge of the battery at time t-1, that is, the historical state, and time t-1 represents a historical time point. SOC(t) represents the set of charge states of n agents at time t.

[0086] Optionally, the above calculation of the income and expenditure information generated by the agent after executing the current behavior at the current time point according to the current price at the current time point in the price information can be specifically implemented through the following steps:

[0087] Determine the current price at the current time point from the price information; determine the electricity cost incurred by the agent after performing the current behavior at the current time point based on the current price, the target increment and the battery capacity of the agent; when the current behavior is a discharge behavior, calculate the battery loss cost incurred by the agent after performing the discharge behavior, and determine the sum of the electricity cost and the battery loss cost as the income and expenditure information generated by the agent after performing the discharge behavior at the current time point; or, when the current behavior is not a discharge behavior, determine the electricity cost as the income and expenditure information.

[0088] It is understandable that the current price at the current time point is determined from the price information, and the different electricity charges generated by the electric vehicle under different actions are calculated based on the current price, target increment and battery capacity, that is, charging will generate charging fees, discharging will receive subsidy fees, and battery replacement will generate battery replacement fees. After determining the electricity fee, it is determined whether the current behavior is a discharge behavior. Discharge behavior will generate battery loss fees. If the current behavior is a discharge behavior, the discharge cost generated by the battery due to discharge loss is calculated, and the sum of the battery loss fee and the discharge subsidy fee is calculated, and the sum is used as the income and expenditure information generated after the intelligent agent executes the discharge behavior; if the current behavior is not a discharge behavior, the electricity fees generated by the vehicle due to different actions are directly used as the income and expenditure information. The income and expenditure information generated under different actions is specifically shown in formula (8).

[0089]

[0090] In the formula, represents the income and expenditure information generated by agent i when executing the current behavior at time t, Indicates electricity cost, C dc Indicates battery depletion cost.

[0091] Among them, the current prices include the first price for charging electric vehicles, the second price for discharging electric vehicles, and the third price for replacing batteries for electric vehicles.

[0092] It is understandable that the economic efficiency of the vehicle charging and discharging process is directly related to the charging and discharging price of the electricity market, where the electricity price changes with the conditions of the electricity market. At time t, the price state of the electricity market is c t , which includes the charging price, discharge subsidy and battery replacement price at this time, as shown in formula (9).

[0093]

[0094] In the formula, c tIndicates the current price of electricity hours at the current time point, represents the electricity price of electric vehicle charging at time t, in units of RMB / kWh, represents the subsidy for electric vehicle discharge at time t, in units of RMB / kWh, It represents the price of replacing the battery of an electric car at time t, in Yuan / kWh.

[0095] Optionally, the above method of determining the electricity cost generated by the agent after performing the current behavior at the current time point based on the current price, the target increment and the battery capacity of the agent can be implemented by the following steps:

[0096] When the current behavior is charging, the electricity cost incurred by the intelligent agent after performing the charging behavior at the current time point is calculated based on the target increment, the battery capacity of the intelligent agent and the first price; or, when the current behavior is discharging, the electricity cost incurred by the intelligent agent after performing the discharging behavior at the current time point is calculated based on the target increment, the battery capacity and the second price; or, when the current behavior is battery replacement, the electricity cost incurred by the intelligent agent after performing the battery replacement behavior at the current time point is calculated based on the battery capacity and the third price; or, when the current behavior is other behaviors, the preset fee is determined as the electricity cost.

[0097] It is understandable that different actions of electric vehicles will generate different electricity charges. Specifically, the current behavior is determined. When the current behavior is charging behavior, the product of the target increment, the first price and the battery capacity is calculated to obtain the charging electricity charge. That is, when the electric vehicle chooses to charge, the generated electricity charge is positive, which means that the owner needs to pay the fee. When the current behavior is discharging behavior, the product of the target increment, the second price and the battery capacity is calculated to obtain the subsidized electricity charge. When the electric vehicle chooses to discharge, the generated discharge subsidy is negative, which means that the owner can obtain the subsidy. When the current behavior is battery replacement behavior, the product of the battery capacity and the third price is calculated to obtain the battery replacement fee. When the electric vehicle chooses to replace the battery, it needs to purchase the power in the entire battery, and the generated electricity charge is positive, which means that the owner needs to pay the battery replacement fee. When the current behavior is other behaviors, the preset fee is determined as the electricity fee, and the preset fee can be 0, that is, when the electric vehicle does not move, no electricity fee is generated. The electricity fee generated under the current behavior is specifically shown in formula (10).

[0098]

[0099] For the specific parameter description in the formula, please refer to the above embodiment and will not be repeated here.

[0100] Optionally, the battery loss cost incurred by the agent after performing the discharge behavior is calculated, which can be achieved through the following steps:

[0101] The discharge depth of the intelligent agent is determined according to the target increment; the remaining number of charge and discharge cycles of the intelligent agent's battery under the discharge depth is calculated according to the preset number of cycles and the preset index, and the preset index is used to characterize the influence of the discharge depth on the charge and discharge cycle; the product of the battery capacity, the remaining number of cycles and the discharge depth is calculated to obtain the total discharge amount of the battery; the ratio of the purchase cost of the battery to the total discharge amount is calculated to obtain the battery loss cost incurred by the intelligent agent after executing the discharge behavior.

[0102] It is understandable that battery loss is mainly related to the discharge depth and cycle life of the battery. The discharge depth (DoD) refers to the ratio of the battery from full charge to discharge to a certain state. It is usually used as an important parameter in the battery use process. The discharge depth of the intelligent agent can be determined according to the target increment, as shown in formula (11). The discharge depth is the change value of the charge state before and after the battery is discharged. After determining the discharge depth, according to the preset number of cycles and the preset index, the remaining number of charge and discharge cycles that the intelligent agent can still perform at the discharge depth is calculated. Among them, the cycle life refers to the number of charge and discharge cycles that the battery can perform before the performance drops to a certain threshold or below a certain threshold. The cycle life has a great relationship with the working intensity and frequency of the battery. The greater the discharge depth, the smaller the cycle life. The relationship between the discharge depth and the cycle life is the discharge characteristic curve of the battery, as shown in formula (12). Subsequently, the product of the battery capacity, the remaining number of cycles and the discharge depth is calculated to obtain the total discharge amount. The total discharge amount can also be understood as the product of the battery cycle life and the single discharge amount of the battery, as shown in formula (13). Then, the ratio of the battery purchase cost to the total discharge amount is calculated to obtain the battery loss cost. The battery loss cost can also be understood as the average cost per unit discharge amount, as shown in formula (14).

[0103] D DoD =soc1-soc2 formula (11)

[0104] Where D DoD Indicates the depth of discharge, soc1 indicates the state of charge value of the battery before discharge, and soc2 indicates the state of charge value of the battery after discharge. When the vehicle battery is discharged, that is, hour,

[0105] L=b·D DoD -d Formula (12)

[0106] Wherein, L represents the number of charge and discharge cycles that the battery can undergo under the depth of discharge (the remaining number of cycles), b is a constant related to the specific battery technology and usage conditions, which represents the theoretical number of cycles when the depth of discharge is 100%, and d is a constant related to the specific battery technology and usage conditions, which represents an index of the degree of influence of the depth of discharge. b and d can usually be obtained by experimental testing.

[0107] E=LE cap D DoD Formula (13)

[0108] Where E represents the total discharge capacity of the battery, in kWh.

[0109]

[0110] In the formula, C dc Indicates the average cost per unit discharge (battery loss cost), unit yuan / kWh, C bat Indicates the purchase cost of the battery, in Yuan.

[0111] It is understandable that the action space of electric vehicles covers a variety of possible behaviors such as charging, discharging, battery replacement and other behaviors. The behavior of agent i at time t is set to According to the setting of the vehicle power supply mode, a set of specific action options is set for each agent at time t, namely, charging, discharging, battery replacement and no action, as shown in formula (15).

[0112]

[0113] In the formula, A(t) represents the action set of n agents at time t, represents the charging of agent i at time t, represents the discharge of agent i at time t, represents agent i changing the battery at time t, Represents that agent i does not take any action (other behaviors) at time t.

[0114] Optionally, the above-mentioned generation of a charging and discharging scheduling strategy for an intelligent agent with the satisfaction of a user corresponding to the intelligent agent as a target based on the price information and the charging and discharging information can be specifically implemented through the following steps:

[0115] Obtain the load rate and / or matching rate corresponding to the intelligent agent, the load rate is used to describe the degree of fluctuation of the power grid load, and the matching rate is used to characterize the degree of matching between the charging demand of electric vehicles and the power generation of new energy; calculate a first value based on the load rate and a preset load fluctuation coefficient; calculate a second value based on the matching rate and a preset new energy consumption coefficient; calculate a third value based on price information, charging and discharging information and a preset user satisfaction coefficient; calculate the sum of the first value, the second value and the third value, and determine the charging and discharging scheduling strategy of the intelligent agent based on the sum.

[0116] Understandably, with the increase in the proportion of renewable energy power generation, especially the large-scale grid connection of wind power generation and photovoltaic power generation, the volatility and intermittency of renewable energy have posed new challenges to the stable operation of the power grid. In response to this problem, this application, based on user attributes and economic considerations, proposes to consider the absorption problem of wind power generation and photovoltaic power generation, and optimizes the charging and discharging strategy of electric vehicles in combination with multiple objectives such as user costs and load fluctuations. Multi-objective optimization problem (Multi-Objective Optimization Problem, MOOP), also known as vector optimization or multi-criteria optimization problem, can be described as: determining a vector composed of decision variables in the feasible domain, which satisfies all constraints and optimizes the vector composed of multiple objective functions. In this type of problem, the optimization process usually needs to consider two or more objectives at the same time, and there is often a certain degree of competition between these objectives, that is, when improving the performance of a certain objective, the performance of other objectives may be impaired. Therefore, the key to multi-objective optimization is to find a balance to achieve the relatively optimal solution of all objectives. Given a set of objective functions f1(x), f2(x),..., f n (x), where x is the decision variable and the goal is to optimize all objective functions simultaneously. Thus, the multi-objective optimization problem can be expressed as formula (16).

[0117] minF(x)=(f1(x),f2(x),...,f n (x)) x∈X Formula (16)

[0118] Where X is the feasible domain of the decision variable.

[0119] Understandably, in addition to considering the battery state of charge of electric vehicles and the power market status, the grid load and renewable energy power generation status are also included, expanding the state space of the intelligent agent to more comprehensively reflect the various factors affecting charging and discharging scheduling.

[0120] Among them, the state set of agent i at time t is shown in formula (17).

[0121]

[0122] This includes the battery charge state, the electricity price state of the power market throughout the day, the power grid load throughout the day, and the new energy output throughout the day, as shown in formula (18).

[0123]

[0124] In the formula, represents the state set of agent i at time t, represents the battery charge state of agent i at time t, c1,c2,c3,...,c t ,...,c m represents the electricity price state of agent i at time t, g1,g2,g3,...,g t ,...,g m represents the load state of agent i at time t, re1, re2, re3, ..., re t ,...,re m Represents the new energy consumption status of agent i at time t.

[0125] Understandably, this multidimensional state representation allows the agent to make the best charging and discharging decision based on comprehensive information, thereby optimizing the energy efficiency and cost of electric vehicles, while taking into account the stability of the power grid and the absorption of new energy output. The action space of the agent includes charging, discharging, battery replacement and no action. The decisions made under different actions also need to consider the absorption of new energy. In order to quantify and optimize these goals, the load rate describing the degree of load fluctuation and the matching rate between vehicle charging demand and new energy power generation are introduced into the reward function. Specifically, the product of the load rate and the preset load fluctuation coefficient is calculated to obtain a first value, the product of the matching rate and the preset new energy absorption coefficient is calculated to obtain a second value, the sum of the income and expenditure information and the current state and the preset user satisfaction coefficient is calculated to obtain a third value, the sum of the first value, the second value and the third value is calculated to obtain the function value of the reward function, and the charging and discharging scheduling strategy is generated according to the function value. The reward function is specifically shown in formula (19).

[0126]

[0127] In the formula, represents the battery state of charge at time t (current state), represents the economic expenditure (income and expenditure information) generated by executing the current behavior, β1 represents the state of charge coefficient, β2 represents the economic expenditure coefficient, which has different values ​​in different time periods, and σ load represents the daily load factor, σ re represents the matching rate, α1, α2, and α3 represent the user satisfaction coefficient, load fluctuation coefficient, and new energy consumption coefficient, respectively, and can be formulated according to actual needs.

[0128] The electric vehicle charging and discharging scheduling strategy generation method provided by the embodiment of the present disclosure effectively improves the accuracy and flexibility of simulating vehicle charging and discharging behaviors through a multi-agent modeling method of electric vehicle charging and discharging that takes into account population attributes; the electric vehicle charging and discharging optimization scheduling method based on multi-agent reinforcement learning enables the model to have excellent real-time and generalization, and when the actual operating environment of the model is complex and changeable, the optimization target can also be dynamically adjusted; in the generation of the charging and discharging scheduling strategy, a multi-objective optimization strategy is adopted, which takes into account the absorption problems of wind power and photovoltaic power generation, comprehensively considers factors such as user costs and grid load fluctuations, improves the utilization rate of new energy, reduces the peak-to-valley difference of grid load, and optimizes the overall energy utilization efficiency.

[0129] Figure 4 The structure diagram of the electric vehicle charging and discharging scheduling strategy generation device provided by the embodiment of the present disclosure is as follows. The charging and discharging scheduling strategy generation device provided by the embodiment of the present disclosure can execute the processing flow provided by the above-mentioned charging and discharging scheduling strategy generation method embodiment, such as Figure 4 As shown, the charging and discharging scheduling strategy generating device 400 includes a first acquiring unit 401, a classification unit 402, a calculation unit 403, a second acquiring unit 404 and a strategy generating unit 405, wherein:

[0130] A first acquisition unit 401 is used to acquire travel data of an electric vehicle cluster and attribute information of a user cluster corresponding to the electric vehicle cluster;

[0131] A classification unit 402, used to classify the electric vehicle cluster into at least one intelligent agent according to the attribute information, wherein the at least one intelligent agent is determined by classifying the electric vehicles with corresponding attributes in the electric vehicle cluster;

[0132] The calculation unit 403 is used to analyze and calculate the target travel data of the intelligent agent and determine the charging and discharging information of each electric vehicle in the intelligent agent;

[0133] The second acquisition unit 404 is used to acquire price information of the electricity market during the period corresponding to the target travel data;

[0134] The strategy generation unit 405 is used to generate a charging and discharging scheduling strategy for the intelligent agent with the satisfaction of the user corresponding to the intelligent agent as the target according to the price information and the charging and discharging information.

[0135] Optionally, the classification unit 402 includes:

[0136] According to the gender and age in the portrait information, target users with the same gender and in the same age range in the user cluster corresponding to the electric vehicle cluster are grouped into one category;

[0137] A target electric vehicle driven by a target user classified into one category is determined in the electric vehicle cluster, and the target electric vehicle is divided into an intelligent agent.

[0138] Optionally, the calculation unit 403 is used for:

[0139] Probabilistic modeling of travel variables in target travel data to generate a travel probability model;

[0140] Based on the travel probability model, the travel behavior of each electric vehicle in the intelligent agent is simulated, and the travel behavior is at least one of charging behavior, discharging behavior, battery replacement behavior and other behaviors;

[0141] The travel behavior of each electric vehicle is analyzed and calculated to determine the charging and discharging information of each electric vehicle.

[0142] The charge and discharge information includes a historical state reflecting the historical battery charge state, a historical time point corresponding to the historical state, and a current behavior reflecting the current charge and discharge behavior, wherein the current time point corresponding to the current behavior is after the historical time point.

[0143] Optionally, the policy generating unit 405 is used to:

[0144] According to the historical state and historical time point, determine the current state of the agent under the current behavior. The current state refers to the battery charge state of the agent at the current time point.

[0145] According to the current price at the current time point in the price information, calculate the income and expenditure information generated by the agent after executing the current behavior at the current time point;

[0146] According to the income and expenditure information and the current state, a charging and discharging scheduling strategy for the agent is generated with the satisfaction of the user corresponding to the agent as the goal.

[0147] Optionally, the policy generating unit 405 is used to:

[0148] When the current behavior is not a battery replacement behavior, determine the target increment of the current behavior of the agent at the current time point for the battery state of charge, and calculate the current state of the agent under the current behavior based on the target increment and the historical state; or

[0149] When the current behavior is a battery replacement behavior, the preset state of charge is determined as the current state.

[0150] Optionally, the policy generating unit 405 is used to:

[0151] In the case where the current behavior is a charging behavior, the charging power for executing the charging behavior is determined according to a charging pile providing charging services for the intelligent agent, and the target increment of the battery state of charge of the charging behavior is calculated according to the battery capacity, charging power and a first duration of the continuous charging behavior starting from the current time point of the intelligent agent; or

[0152] In the case where the current behavior is a discharging behavior, the discharge power for executing the discharging behavior is determined according to the charging pile, and the target increment of the battery state of charge due to the discharging behavior is calculated according to the battery capacity, the discharge power and the second duration of the continuous discharging behavior starting from the current time point; or,

[0153] When the current behavior is other behaviors, the preset value is determined as the target increment of the other behaviors with respect to the battery state of charge.

[0154] Optionally, the policy generating unit 405 is used to:

[0155] If the historical time point is the initial time point when the agent initially participates in scheduling, the current state is calculated based on the target increment and the initial state of the agent at the initial time point, wherein the historical time point is after the initial time point, and the initial state refers to the remaining charge state of the agent; or,

[0156] If the historical time point is not the initial time point, the current state is calculated based on the target increment and the historical state.

[0157] Optionally, the policy generating unit 405 is used to:

[0158] Determine the current price at the current time from the price information;

[0159] Determine the electricity cost of the agent after performing the current behavior at the current time point based on the current price, target increment, and the agent's battery capacity;

[0160] When the current behavior is a discharge behavior, the battery loss cost generated by the intelligent agent after performing the discharge behavior is calculated, and the sum of the electricity cost and the battery loss cost is determined as the income and expenditure information generated after the intelligent agent performs the discharge behavior at the current time point; or, when the current behavior is not a discharge behavior, the electricity cost is determined as the income and expenditure information.

[0161] Among them, the current prices include the first price for charging electric vehicles, the second price for discharging electric vehicles, and the third price for replacing batteries for electric vehicles.

[0162] Optionally, the policy generating unit 405 is used to:

[0163] In the case where the current behavior is a charging behavior, the electricity fee generated by the agent after performing the charging behavior at the current time point is calculated according to the target increment, the battery capacity of the agent and the first price; or,

[0164] In the case where the current behavior is a discharge behavior, the electricity cost incurred by the agent after performing the discharge behavior at the current time point is calculated based on the target increment, the battery capacity and the second price; or,

[0165] In the case where the current behavior is battery replacement, the electricity fee generated by the agent after performing the battery replacement behavior at the current time point is calculated based on the battery capacity and the third price; or,

[0166] In the case where the current behavior is other behaviors, the preset fee is determined as the electricity fee.

[0167] Optionally, the policy generating unit 405 is used to:

[0168] Determine the agent's discharge depth based on the target increment;

[0169] Calculate the remaining number of charge and discharge cycles of the intelligent body's battery at the depth of discharge according to the preset number of cycles and the preset index, where the preset index is used to characterize the degree of influence of the depth of discharge on the charge and discharge cycle;

[0170] Calculate the product of battery capacity, remaining cycle number and discharge depth to obtain the total discharge capacity of the battery;

[0171] Calculate the ratio of the battery purchase cost to the total discharge amount to obtain the battery loss cost incurred by the agent after performing the discharge behavior.

[0172] Optionally, the policy generating unit 405 is used to:

[0173] Obtain the load rate and / or matching rate corresponding to the intelligent agent. The load rate is used to describe the degree of load fluctuation of the power grid, and the matching rate is used to characterize the matching degree between the charging demand of electric vehicles and the power generation of new energy.

[0174] A first value is obtained by calculating according to the load rate and a preset load fluctuation coefficient;

[0175] A second value is calculated based on the matching rate and the preset new energy consumption coefficient;

[0176] A third value is calculated based on the price information, the charging and discharging information, and a preset user satisfaction coefficient;

[0177] A sum of the first value, the second value, and the third value is calculated, and a charging and discharging scheduling strategy of the intelligent agent is determined based on the sum.

[0178] Figure 4The electric vehicle charging and discharging scheduling strategy generation device of the illustrated embodiment can be used to execute the technical solution of the above-mentioned method embodiment. Its implementation principle and technical effect are similar and will not be repeated here.

[0179] Figure 5 This is a schematic diagram of the structure of an electronic device provided by an embodiment of the present disclosure. Figure 5 , which shows a schematic diagram of the structure of an electronic device 500 suitable for implementing the embodiment of the present disclosure. The electronic device 500 in the embodiment of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle terminals (such as vehicle navigation terminals), wearable electronic devices, etc., and fixed terminals such as digital TVs, desktop computers, smart home devices, etc. Figure 5 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0180] like Figure 5 As shown, the electronic device 500 may include a processing device 501 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage device 508 into a random access memory (RAM) 503 to implement the electric vehicle charging and discharging scheduling strategy generation method of the embodiment described in the present disclosure. In RAM 503, various programs and data required for the operation of the electronic device 500 are also stored. The processing device 501, ROM 502, and RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0181] Typically, the following devices may be connected to the I / O interface 505: an input device 506 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 507 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 508 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 509. The communication device 509 may allow the electronic device 500 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 5 The electronic device 500 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0182] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains a program code for executing the method shown in the flowchart, thereby realizing the electric vehicle charging and discharging scheduling strategy generation method as described above. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 509, or installed from the storage device 508, or installed from the ROM 502. When the computer program is executed by the processing device 501, the above-mentioned functions defined in the method of the embodiment of the present disclosure are executed.

[0183] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0184] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0185] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0186] Optionally, when the above one or more programs are executed by the electronic device, the electronic device may also execute other steps described in the above embodiments.

[0187] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0188] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0189] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware, wherein the name of a unit does not, in some cases, limit the unit itself.

[0190] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0191] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0192] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or gateway that includes a series of elements includes not only those elements, but also other elements that are not explicitly listed, or also includes elements inherent to such process, method, article or gateway. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or gateway that includes the elements.

[0193] The above description is only a specific embodiment of the present disclosure, so that those skilled in the art can understand or implement the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for generating a charging and discharging scheduling strategy for an electric vehicle, characterized in that: include: Acquire travel data of an electric vehicle cluster and attribute information of a user cluster corresponding to the electric vehicle cluster; Dividing the electric vehicle cluster into at least one intelligent agent according to the attribute information, wherein the at least one intelligent agent is determined by classifying the electric vehicles with corresponding attributes in the electric vehicle cluster; Analyze and calculate the target travel data of the intelligent agent to determine the charging and discharging information of each electric vehicle in the intelligent agent; the charging and discharging information includes a historical state reflecting the historical battery charge state, a historical time point corresponding to the historical state, and a current behavior reflecting the current charging and discharging behavior, and the current time point corresponding to the current behavior is after the historical time point; Obtain price information of the electricity market during the period corresponding to the target travel data; Generating a charging and discharging scheduling strategy for the agent with the satisfaction of the user corresponding to the agent as the target according to the price information and the charging and discharging information, including: determining the current state of the agent under the current behavior according to the historical state and the historical time point, wherein the current state refers to the battery charge state of the agent at the current time point; calculating the income and expenditure information generated by the agent after executing the current behavior at the current time point according to the current price at the current time point in the price information; generating a charging and discharging scheduling strategy for the agent with the satisfaction of the user corresponding to the agent as the target according to the income and expenditure information and the current state; Wherein, determining the current state of the agent under the current behavior according to the historical state and the historical time point includes: when the current behavior is not a battery replacement behavior, determining a target increment of the current behavior of the agent at the current time point for the battery state of charge, and calculating the current state of the agent under the current behavior according to the target increment and the historical state; Wherein, determining the target increment of the battery state of charge of the current behavior of the intelligent body at the current time point includes: when the current behavior is a charging behavior, determining the charging power for executing the charging behavior according to a charging pile providing charging service for the intelligent body, and calculating the target increment of the battery state of charge of the charging behavior according to the battery capacity of the intelligent body, the charging power and a first duration of the charging behavior starting from the current time point.

2. The method according to claim 1, characterized in that The analyzing and calculating the target travel data of the intelligent agent to determine the charging and discharging information of each electric vehicle in the intelligent agent includes: Probabilistically modeling the travel variables in the target travel data to generate a travel probability model; Simulating the travel behavior of each electric vehicle in the intelligent agent based on the travel probability model, wherein the travel behavior is at least one of charging behavior, discharging behavior, battery replacement behavior and other behaviors; The travel behaviors of the electric vehicles are analyzed and calculated to determine the charging and discharging information of the electric vehicles.

3. The method according to claim 1, characterized in that The determining the current state of the agent under the current behavior according to the historical state and the historical time point also includes: In a case where the current behavior is the battery replacement behavior, a preset state of charge is determined as the current state.

4. The method according to claim 1, characterized in that: The determining of the target increment of the battery state of charge of the current behavior of the agent at the current time point also includes: In the case where the current behavior is a discharging behavior, the discharging power for executing the discharging behavior is determined according to the charging pile, and the target increment of the battery state of charge for the discharging behavior is calculated according to the battery capacity, the discharging power and a second duration of the discharging behavior starting from the current time point; or In a case where the current behavior is other behaviors, a preset value is determined as a target increment of the other behaviors for the battery state of charge.

5. The method according to claim 1, characterized in that The step of calculating the current state of the agent under the current behavior according to the target increment and the historical state includes: If the historical time point is the initial time point when the agent initially participates in scheduling, the current state is calculated according to the target increment and the initial state of the agent at the initial time point, wherein the historical time point is after the initial time point, and the initial state refers to the remaining charge state of the agent; or If the historical time point is not the initial time point, the current state is calculated based on the target increment and the historical state.

6. The method according to claim 1, characterized in that The calculating, according to the current price at the current time point in the price information, the income and expenditure information generated after the agent performs the current behavior at the current time point includes: Determining the current price at the current time point from the price information; Determine, based on the current price, the target increment, and the battery capacity of the agent, the electricity cost incurred by the agent after performing the current behavior at the current time point; In the case where the current behavior is a discharging behavior, the battery loss cost generated by the intelligent agent after performing the discharging behavior is calculated, and the sum of the electricity cost and the battery loss cost is determined as the income and expenditure information generated by the intelligent agent after performing the discharging behavior at the current time point; or, in the case where the current behavior is not a discharging behavior, the electricity cost is determined as the income and expenditure information.

7. The method according to claim 6, characterized in that The current price includes a first price for charging the electric vehicle, a second price for discharging the electric vehicle, and a third price for replacing the battery of the electric vehicle. The determining, based on the current price, the target increment, and the battery capacity of the agent, of the electricity fee generated after the agent performs the current behavior at the current time point includes: In the case where the current behavior is a charging behavior, the electricity fee generated after the intelligent agent performs the charging behavior at the current time point is calculated according to the target increment, the battery capacity of the intelligent agent and the first price; or In the case where the current behavior is a discharging behavior, the electricity fee generated by the intelligent agent after performing the discharging behavior at the current time point is calculated according to the target increment, the battery capacity and the second price; or In the case where the current behavior is a battery replacement behavior, the electricity fee generated by the intelligent agent after performing the battery replacement behavior at the current time point is calculated according to the battery capacity and the third price; or In a case where the current behavior is other behaviors, the preset fee is determined as the electricity fee.

8. The method according to claim 6, characterized in that The calculating the battery loss cost generated by the intelligent agent after executing the discharging behavior includes: determining a discharge depth of the agent according to the target increment; Calculating the remaining number of charge and discharge cycles of the battery of the intelligent body at the depth of discharge according to the preset number of cycles and the preset index, wherein the preset index is used to characterize the influence of the depth of discharge on the charge and discharge cycle; Calculating the product of the battery capacity, the remaining number of cycles and the depth of discharge to obtain a total discharge capacity of the battery; The ratio of the purchase cost of the battery to the total discharge amount is calculated to obtain the battery loss cost generated by the intelligent agent after executing the discharge behavior.

9. The method according to claim 1, characterized in that: The step of generating a charging and discharging scheduling strategy for the agent with the satisfaction of the user corresponding to the agent as the target according to the price information and the charging and discharging information comprises: Obtaining a load rate and / or matching rate corresponding to the intelligent agent, wherein the load rate is used to describe the degree of fluctuation of the load on the power grid, and the matching rate is used to characterize the degree of matching between the charging demand of the electric vehicle and the power generation of new energy; Calculate a first value according to the load rate and a preset load fluctuation coefficient; A second value is calculated based on the matching rate and a preset new energy consumption coefficient; Calculate a third value according to the price information, the charging and discharging information, and a preset user satisfaction coefficient; A sum of the first value, the second value, and the third value is calculated, and a charging and discharging scheduling strategy of the intelligent agent is determined based on the sum.

Images