An electric vehicle cluster multi-set meal master-slave game ordered charging optimization method

Abstract

The application discloses a kind of electric vehicle cluster multi-set meal master-slave game ordered charging optimization method, it is related to charging optimization technical field, including the following steps, acquisition charging information, based on probability prediction method fitting the probability distribution of different charging piles each period charging demand;Through the probability distribution of charging demand and distribution network purchase and sale electricity price, set electric vehicle charging package;According to charging information, construct charging cost model, combine electric vehicle charging package, form electric vehicle charging load package selection method.The application considers road network information, driver behavior trajectory and power grid topology structure information by comprehensively, fitting the probability distribution of different charging piles each period charging demand based on improved LightGBM model method, and according to electric vehicle individualized charging demand, formulate fixed price package mode, peak-valley price package mode and ladder price package mode form, and electric vehicle charging load is selected according to own charging demand.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of charging optimization, and in particular to a multi-set meal master-slave game ordered charging optimization method for electric vehicle clusters. BACKGROUND

[0002] With the decrease of traditional fossil energy and the increasing environmental pollution, large-scale development of new energy has become the core direction of China's energy and power industry. The proposed double-carbon strategy further promotes the development of new energy, and the installed capacity of renewable energy such as wind power and photovoltaic is rising. While giving the power system a green attribute, it has intensified the uncertainty of safe and stable operation. Electric vehicles are considered as "new energy" on the load side. Compared with traditional fuel vehicles, electric vehicles realize operation through battery charging and discharging, effectively responding to the national double-carbon strategy.

[0003] The number of electric vehicles in China continues to rise, and large-scale electric vehicle charging loads will bring a series of challenges and opportunities to the distribution network. Electric vehicle charging loads are affected by factors such as the number of vehicles, types, charging time, charging duration, and charging frequency, and have randomness in space-time distribution. When charging is disordered, the peak of charging load often overlaps with the peak of residential electricity load, resulting in an increase in peak-valley difference of electricity load, increasing the peak regulation pressure of the power grid, and reducing the investment benefit of the distribution network. With the large-scale access of new energy power sources, their volatility and randomness are superimposed with the randomness of charging load, making power system dispatching more complex. While electric vehicle ordered charging has positive effects such as peak load reduction and clean energy consumption promotion, it can effectively alleviate the peak regulation pressure of the power grid.

[0004] Electricity price control is an effective means to guide electric vehicle ordered charging, which uses time-of-use electricity price, real-time electricity price, and peak electricity price to increase charging costs when the load is high and reduce charging costs when the load is low, and uses the users' psychology of saving electricity bills to guide users to charge during the load valley period. At the same time, the real-time electricity price and the tiered electricity price mechanism can also guide the user's demand period to coincide with the period when new energy generation is sufficient, which is in line with the intermittent characteristics of new energy output, reduces the phenomenon of abandoned wind and light, improves the reliability of the power grid, and maintains the balance between supply and demand in the electricity market.

[0005] Residential communities are areas with high-frequency electric vehicle charging, and face many problems such as a surge in charging load and limited transformer capacity, resulting in conflicts between residential electricity and electric vehicle charging demand. It is necessary to guide electric vehicle ordered charging to alleviate the problem of tight power supply and demand in residential communities, which is conducive to delaying the high cost of replacing transformers in residential communities and further meeting the individual charging needs of different electric vehicle users. SUMMARY

[0006] In view of the problems existing in the existing electric vehicle cluster multi-set meal master-slave game orderly charging optimization method, the present application is proposed.

[0007] Therefore, the problem to be solved by the present application is that the electric vehicle charging time cannot be guided, which will cause the peak-valley difference of power consumption load to increase, and increase the peak regulation pressure of the power grid.

[0008] To solve the above technical problems, the present application provides the following technical solutions:

[0009] In the first aspect, the present application embodiment provides an electric vehicle cluster multi-set meal master-slave game orderly charging optimization method, which includes the following steps,

[0010] Collect charging information, and fit the probability distribution of charging demand of different charging piles in each period based on a probability prediction method;

[0011] Set an electric vehicle charging set meal through the probability distribution of charging demand and the buying and selling electricity price of the distribution network;

[0012] According to the charging information, a charging cost model is constructed, and an electric vehicle charging load set meal selection method is formed in combination with the electric vehicle charging set meal.

[0013] As a preferred scheme of the electric vehicle cluster multi-set meal master-slave game orderly charging optimization method, the charging information includes community surrounding road network information, vehicle owner behavior trajectory, and power grid topology structure;

[0014] The community surrounding road network information includes road network nodes, road connection conditions, road grades, real-time traffic, and congestion conditions;

[0015] The vehicle owner behavior trajectory includes travel time probability distribution, road section driving time, destination stay time, trip departure time, user power state, and charging duration;

[0016] The power grid topology structure includes transformer capacity, community different branch parameters, node position, and node parameters.

[0017] As a preferred scheme of the electric vehicle cluster multi-set meal master-slave game orderly charging optimization method, the probability distribution fitting method of charging demand of different charging piles in each period includes,

[0018] The single-point prediction objective function is set based on the improved LightGBM model, and the mathematical formula is represented as:

[0019]

[0020] In the formula, L represents the objective function of the LightGBM model, Let be the predicted value of the charging pile in the t-th time period. Compared with the true value x t The error relationship between them, Ω(f) k ) represents the regularization term of the LightGBM model, γ and γ are both penalty coefficients, A represents the number of nodes in the k-th tree of the LightGBM model, and ω a This represents the weight of the a-th node in the k-th tree of the LightGBM model;

[0021] The objective function is derived by taking the derivative of the i-th iteration using the forward distribution algorithm, and the number of weights ω is then calculated. a The objective function is obtained by finding the minimum value of , and the specific formula is as follows:

[0022]

[0023] I a ={i|q(x i )=a}

[0024] In the formula, L (i) Let g be the objective function after the derivative of the i-th iteration of the LightGBM model, I represent the total number of iterations of the LightGBM model, ω represent the weight of the k-th tree in the LightGBM model, and g i and h i Let f represent the coefficients of the derivative in the i-th iteration, and f and f are both penalty coefficients. i (x i ) represents the derivative of the actual values ​​of the charging pile in the LightGBM model for each time period in the i-th iteration, I a Let ω represent the set of nodes in each tree, and let ω represent the weight of the k-th tree in the LightGBM model after the i-th iteration.

[0025] A nonparametric kernel density estimation model is used to establish a probability distribution model for the charging demand prediction error of charging piles in different time periods. The mathematical formula is as follows:

[0026]

[0027] In the formula, denoted as x, where x′ represents the predicted value of the charging pile for each time period; and x′ represents the predicted error value of the charging pile for each time period. Let σ represent the probability distribution model of the charging demand prediction error for each time period, where σ represents the free parameters of the probability distribution model.

[0028] As a preferred embodiment of the electric vehicle cluster multi-package master-slave game-based ordered charging optimization method described in this invention, wherein: the electric vehicle charging packages include a fixed electricity price package mode, a peak-valley electricity price package mode, and a tiered electricity price package mode;

[0029] The fixed electricity price package mode is a fixed electricity price all day long without charging capacity limit.

[0030] The peak-valley electricity price package mode is to set up peak-valley electricity price to guide electric vehicle charging load and participate in peak shaving and valley filling of the power grid.

[0031] The stepped electricity price package mode sets the charging capacity of the electric vehicle charging load as N grades, and different electricity quantity standards and electricity price standards are implemented in each grade of the stepped incremental electricity price to guide the agent electric vehicle charging load to reasonably charge.

[0032] As a preferred scheme of the electric vehicle cluster multi-package master-slave game ordered charging optimization method, in the fixed electricity price package mode, the electricity selling income of the mth charging pile is represented as:

[0033]

[0034] In the formula, is the fixed electricity price package mode income of the charging pile in the dth day period t, is the fixed electricity price of the charging pile in the dth day period t, is the electricity selling quantity of the charging pile in the dth day period t;

[0035] In the peak-valley electricity price package mode, the electricity selling income of the mth charging pile is represented as:

[0036]

[0037]

[0038] In the formula, is the peak-valley electricity price package mode income of the charging pile in the dth day period t, is the set peak-valley flat period electricity price, P , T M and T V is the set peak-valley flat period, and is the settlement price of the charging pile in the peak-valley flat period;

[0039] In the stepped electricity price package mode, the electricity selling income of the mth charging pile is represented as:

[0040]

[0041] In the formula, is the stepped electricity price package mode income in the dth day period t, is the stepped electricity price of the charging pile in the period t, and Respectively, the electricity price of the charging pile under different electricity standards, Respectively, the electricity price of the charging pile under different electricity standards, And Respectively, the electricity price of the charging pile under different electricity standards,

[0042] As a preferred solution of the electric vehicle cluster multi-suite master-slave game ordered charging optimization method of the application, wherein: the charging cost model includes a power purchase cost model, an investment and construction cost model, and an operation and maintenance cost model, wherein the power purchase cost model is represented by a mathematical formula as:

[0043]

[0044] The investment and construction cost model is represented by a mathematical formula as:

[0045]

[0046] The operation and maintenance cost model is represented by a mathematical formula as:

[0047]

[0048] Through the charging cost model, the charging pile economic benefit model is calculated, which is specifically represented as

[0049]

[0050] F up =I ev -C inv -C yw -C gs

[0051] In the formula, F up represents the economic benefit of the charging pile, I ev represents the charging pile package electricity sales revenue, C inv represents the charging pile investment and construction cost, C yw represents the charging pile operation and maintenance cost, C gs represents the charging pile power purchase cost, M represents the total number of charging piles, ψ E represents the unit capacity charging pile investment and construction cost, γ represents the discount rate of all charging piles of the charging pile, represents the charging pile life cycle of the mth charging pile, represents the rated capacity of the mth charging pile, represents the unit capacity operation and maintenance cost of the mth charging pile, represents the power purchase power of the charging pile to the power distribution network on the dth day and t period, The power selling price of the power grid at the charging pile in the dth time period t is represented as P d t ;

[0052] The charging cost of the electric vehicle charging load is represented as C

[0053] F down =C ev +C loss

[0054] The charging cost includes the package charging cost and the loss cost, and the package charging cost is represented as C

[0055]

[0056] The loss cost is represented as C

[0057]

[0058] In the formula, F down represents the charging cost of the electric vehicle charging load, C ev represents the package charging cost of the electric vehicle, C loss represents the loss cost of the electric vehicle, k1, k2 and k3 are all 0-1 parameters, and satisfy k1+k2+k3=1, i represents the i th charging package of the charging pile, and λ represents the unit charging power cost loss of the electric vehicle charging load.

[0059] As a preferred scheme of the electric vehicle cluster multi-package master-slave game ordered charging optimization method, the electric vehicle charging load package selection method comprises the following steps: adopting a master-slave game model to depict the power interaction and dynamic game between the charging pile and the electric vehicle charging load, and obtaining a Nash equilibrium solution by nesting a CPLEX commercial solver through a cuckoo optimization algorithm.

[0060] The master-slave game model between the charging pile and the electric vehicle charging load is represented as:

[0061] G={N;S 充电桩 ;S 充电负荷 ;F up ;F down}

[0062] In the formula, N represents a participant, and its set is N={charging pile U electric vehicle charging load};

[0063] S 充电桩 represents the strategy of the master charging pile, and its set is:

[0064]

[0065] S 充电负荷 represents the strategy of the slave electric vehicle charging load, and its set is:

[0066]

[0067] In the formula, G represents a game model from the game;

[0068] Finally, the obtained Nash equilibrium solution forms the charging pile package pricing result and the electric vehicle charging load package selection method.

[0069] In a second aspect, the embodiments of the present application provide an electric vehicle cluster multi-package master-slave game orderly charging optimization system, which comprises a charging data acquisition module, a charging package design module, a cost model construction module, and an optimization decision module.

[0070] The charging data acquisition module acquires real-time data related to charging, including charging demand of electric vehicles, location information of vehicles, state of power grids, and electricity price information.

[0071] The charging package design module designs multiple charging packages for electric vehicle users to select based on the purchase and sale electricity prices of power grids and the probability distribution of charging demand.

[0072] The cost model construction module is used to estimate and analyze the cost of the entire charging system, including the purchase electricity cost, infrastructure construction cost, and operation and maintenance cost.

[0073] The optimization decision module adopts a master-slave game model in game theory and a cuckoo search algorithm embedded with a commercial solver to provide optimization decisions for charging pile pricing and charging load management.

[0074] In a third aspect, the embodiments of the present application provide a computer device comprising a memory and a processor, and the memory stores a computer program, wherein the processor implements any step of the electric vehicle cluster multi-package master-slave game orderly charging optimization method described above when executing the computer program.

[0075] In a fourth aspect, the embodiments of the present application provide a computer readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement any step of the electric vehicle cluster multi-package master-slave game orderly charging optimization method described above.

[0076] The application has the beneficial effects that: by comprehensively considering the road network information, the vehicle owner behavior trajectory and the power grid topology structure information, the probability distribution of the charging demand of each period of different charging piles is fitted based on the improved LightGBM model method, and the fixed electricity price package mode, the peak-valley electricity price package mode and the ladder electricity price package mode are formulated according to the individualized charging demand of the electric vehicles, the electric vehicle charging load selects the package according to the charging demand, the charging pile and the electric vehicle charging load form the master-slave game based on the autonomy and the profit-seeking, the different package prices of the charging pile and the electric vehicle charging load package selection results are formed through the interactive game and repeated iteration, the individualized charging demand of the electric vehicle charging load is met while the charging cost is reduced, and it is beneficial to guide the orderly charging of the electric vehicles to relieve the power supply tension in the community in some periods. BRIEF DESCRIPTION OF DRAWINGS

[0077] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor. Among them:

[0078] Figure 1 The flowchart of the electric vehicle cluster multi-package master-slave game orderly charging optimization method. DETAILED DESCRIPTION

[0079] In order to make the above-mentioned purposes, features and advantages of the present application more apparent and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0080] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the connotation of the present application, therefore the present application is not limited to the specific embodiments disclosed below.

[0081] Secondly, the "one embodiment" or "embodiment" referred to herein means that the specific features, structures or characteristics can be included in at least one implementation of the present application. "In one embodiment" appearing in different places in the specification does not mean the same embodiment, nor is it an independent or alternative embodiment that excludes other embodiments.

[0082] The application is described in detail in combination with the schematic diagram. In the detailed description of the embodiments of the application, the cross-sectional view of the device structure is partially enlarged without the general proportion for the convenience of illustration, and the schematic diagram is only an example, which should not limit the scope of protection of the application herein. In addition, the three-dimensional spatial dimensions of length, width and depth should be included in actual production.

[0083] Meanwhile, in the description of the application, it should be noted that the terms "upper, lower, inner and outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the application. In addition, the terms "first, second or third" are only for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0084] Unless otherwise specifically defined and limited in the application, the terms "mounting, connecting, connecting" should be broadly understood, for example: it can be fixedly connected, detachably connected or integrally connected; it can also be mechanically connected, electrically connected or directly connected, it can also be indirectly connected through an intermediate medium, or it can be the communication between two elements. For those skilled in the art, the specific meaning of the above terms in the application can be understood according to the specific circumstances.

[0085] Embodiment 1

[0086] Reference Figure 1 For the first embodiment of the application, the embodiment provides a multi-set meal master-slave game ordered charging optimization method for electric vehicle clusters, comprising the following steps,

[0087] S1, collect charging information, and fit the probability distribution of charging demand of different charging piles in each period based on a probability prediction method.

[0088] The charging information includes community surrounding road network information, vehicle owner behavior trajectory, and power grid topology structure;

[0089] Among them, the community surrounding road network information includes road network nodes, road connection conditions, road grades, real-time traffic, and congestion conditions;

[0090] The vehicle owner behavior trajectory includes travel time probability distribution, road section driving time, destination stay time, trip departure time, user power state, and charging duration;

[0091] The power grid topology structure includes transformer capacity, community different branch parameters, node position, and node parameters.

[0092] The probability distribution fitting method of charging demand of different charging piles in each period includes,

[0093] The single-point prediction objective function is set based on the improved LightGBM model, and the mathematical formula is as follows:

[0094]

[0095]

[0096] In the formula, L represents the objective function of the LightGBM model, represents the charging pile prediction value of the tth time period , and x represents the error relationship between the true value t , Ω(f k ) represents the regular term of the LightGBM model, which is used to optimize the model parameters, γ and γ both represent the penalty coefficient, A represents the number of nodes of the kth tree of the LightGBM model, and ω a represents the weight of the ath node of the kth tree of the LightGBM model.

[0097] The forward distribution algorithm is used to derive the objective function after the ith iteration, to solve the minimum value of the weight ω a , and obtain the objective function, and the specific formula is as follows:

[0098]

[0099] In the formula, L (i) represents the objective function of the LightGBM model after the ith iteration, I represents the total number of iterations of the LightGBM model, ω represents the weight of the kth tree of the LightGBM model, g i and h i respectively represent the coefficients of the ith iteration and are both penalty coefficients, f i (x i ) represents the actual value of each time period of the charging pile after the ith iteration of the LightGBM model, I a represents the set of nodes of each tree, and ω represents the weight of the kth tree of the LightGBM model after the ith iteration.

[0100] A non-parametric kernel density estimation model is used to establish a probability distribution model for the prediction error of the charging demand of each time period of the charging pile, and the mathematical formula is as follows:

[0101]

[0102] In the formula, represents the prediction value of each time period of the charging pile, x' represents the prediction error value of each time period of the charging pile, represents the probability distribution model of the prediction error of the charging demand of each time period, and σ represents a free parameter of the probability distribution model.

[0103] S2, setting an electric vehicle charging package through a probability distribution of charging demand and a power grid buying and selling electricity price.

[0104] The electric vehicle charging package includes a fixed electricity price package mode, a peak-valley electricity price package mode, and a ladder electricity price package mode.

[0105] The fixed electricity price package mode is a fixed electricity price throughout the day without charging quantity limit.

[0106] The peak-valley electricity price package mode is to formulate a peak-valley electricity price to guide electric vehicle charging load and participate in power grid peak shaving and valley filling.

[0107] The ladder electricity price package mode sets the electric vehicle charging load charging quantity as N grades, implements different electricity quantity standards and electricity price standards in each grade of the ladder increasing electricity price, and guides the agent electric vehicle charging load to reasonably charge.

[0108] In the fixed electricity price package mode, the electricity selling revenue of the mth charging pile is represented as:

[0109]

[0110] In the formula, represents the fixed electricity price package mode revenue of the charging pile in the tth time period of the dth day, represents the fixed electricity price of the charging pile in the tth time period of the dth day, represents the electricity selling quantity of the charging pile in the tth time period of the dth day;

[0111] In the peak-valley electricity price package mode, the electricity selling revenue of the mth charging pile is represented as:

[0112]

[0113]

[0114] In the formula, represents the peak-valley electricity price package mode revenue of the charging pile in the tth time period of the dth day, represents the formulated peak-valley flat time period electricity price, P represents the formulated peak-valley flat time period electricity price, M represents the formulated peak-valley flat time period electricity price, V represents the formulated peak-valley flat time period, and represent the settlement price of the charging pile in the peak-valley flat time period;

[0115] In the ladder electricity price package mode, the electricity selling revenue of the mth charging pile is represented as:

[0116]

[0117] In the formula, This represents the revenue from the tiered electricity pricing package during time period t on day d. This represents the tiered electricity price for the charging station during time period t. and These represent the electricity prices for charging stations under different electricity consumption standards. This represents the electricity sold by the charging station during time period t on day d, according to the tiered electricity pricing package. and This represents the boundary between different electricity standards.

[0118] S3. Construct a charging cost model based on charging information, and combine it with electric vehicle charging packages to form a method for selecting electric vehicle charging load packages.

[0119] The charging cost model includes an electricity purchase cost model, an investment and construction cost model, and an operation and maintenance cost model. The electricity purchase cost model is expressed mathematically as follows:

[0120]

[0121] The investment and construction cost model can be expressed mathematically as follows:

[0122]

[0123] The operation and maintenance cost model is expressed mathematically as follows:

[0124]

[0125] The economic benefit model of charging piles is calculated using a charging cost model, specifically expressed as follows:

[0126]

[0127] F up =I ev -C inv -C yw -C gs

[0128] In the formula, F up Represented as the economic benefits of charging stations, I ev C represents the revenue from electricity sales for charging station packages. inv C represents the investment and construction cost of charging stations. yw C represents the operation and maintenance cost of the charging station. gs ψ represents the cost of electricity purchased from the power grid by the charging pile, M represents the total number of charging piles, and ψ E γ represents the investment and construction cost per unit capacity of the charging pile, and γ represents the discount rate for all charging piles. This represents the lifespan of the m-th charging pile. This represents the rated capacity of the m-th charging pile. This represents the unit capacity operation and maintenance cost of the m-th charging pile. This represents the power purchased by the charging pile from the distribution network during time period t on day d. This represents the electricity price sold by the charging pile on the distribution network during time period t on day d.

[0129] The formula for calculating the charging cost of electric vehicles is expressed as follows:

[0130] F down =C ev +C loss

[0131] The charging cost includes the cost of the package charging and the cost of lost charges. The cost of the package charging is expressed as follows:

[0132]

[0133] The cost of loss is expressed as:

[0134]

[0135] In the formula, F down C represents the charging cost of an electric vehicle charging load. ev C represents the charging cost of an electric vehicle package. loss Let λ represent the loss cost of electric vehicles, k1, k2, and k3 are all 0-1 parameters, and satisfy k1+k2+k3=1, i represents the i-th charging package of the charging pile, and λ represents the unit charging power cost loss of the electric vehicle charging load.

[0136] The method for selecting electric vehicle charging load packages includes: using a master-slave game model to characterize the power interaction and dynamic game between charging piles and electric vehicle charging loads, and obtaining the Nash equilibrium solution by nesting the CPLEX commercial solver with the Cuckoo optimization algorithm;

[0137] The master-slave game model between charging piles and electric vehicle charging loads can be represented as follows:

[0138] G = {N; S} 充电桩 S 充电负荷 ;F up ;F down}

[0139] Where N represents the participants, and their set is N = {charging piles ∪ electric vehicle charging loads};

[0140] S 充电桩 The main charging pile strategy is represented by the following set:

[0141]

[0142] S 充电负荷 The charging load strategy of the electric vehicle is represented as G, and the set is:

[0143]

[0144] In the formula, G represents a game model;

[0145] Finally, the obtained Nash equilibrium solution forms the charging pile package pricing result and the electric vehicle charging load package selection method.

[0146] In summary, by comprehensively considering the road network information, the behavior trajectory of the vehicle owner, and the topology information of the power grid, the improved LightGBM model method is used to fit the probability distribution of the charging demand of each period of different charging piles, and according to the individualized charging demand of the electric vehicle, the fixed electricity price package mode, the peak-valley electricity price package mode, and the ladder electricity price package mode are formulated. The electric vehicle charging load selects the package according to its own charging demand. The charging pile and the electric vehicle charging load form a master-slave game based on autonomy and profit-seeking. Through interactive game and repeated iteration, the different package prices of the charging pile and the electric vehicle charging load package selection results are formed. While meeting the individualized charging demand of the electric vehicle charging load, the charging cost is reduced, which is conducive to guiding the orderly charging of the electric vehicle to alleviate the tight power supply in some periods of the community.

[0147] Embodiment 2

[0148] On the basis of the first embodiment, the embodiment further provides an electric vehicle cluster multi-package master-slave game orderly charging optimization system, which includes a charging data acquisition module, a charging package design module, a cost model construction module, and an optimization decision module.

[0149] The charging data acquisition module acquires real-time data related to charging, including the charging demand of the electric vehicle, the position information of the vehicle, the state of the power grid, and the electricity price information.

[0150] The charging package design module designs multiple charging packages based on the purchase and sale electricity price of the power grid and the probability distribution of the charging demand for the electric vehicle user to select.

[0151] The cost model construction module is used to estimate and analyze the cost of the entire charging system, including the purchase electricity cost, the infrastructure construction cost, and the operation and maintenance cost.

[0152] The optimization decision module uses the master-slave game model in game theory and the cuckoo search algorithm embedded in the commercial solver to provide optimization decisions for the pricing of the charging pile and the management of the charging load.

[0153] The embodiment also provides a computer device suitable for the electric vehicle cluster multi-set meal master-slave game orderly charging optimization method, which comprises a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to realize the electric vehicle cluster multi-set meal master-slave game orderly charging optimization method proposed in the above embodiment.

[0154] The computer device can be a terminal, which comprises a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0155] The embodiment also provides a storage medium having a computer program stored thereon, which is executed by a processor to realize the electric vehicle cluster multi-set meal master-slave game orderly charging optimization method proposed in the above embodiment.

[0156] The storage medium proposed in the embodiment belongs to the same inventive concept as the data storage method proposed in the above embodiment, and the technical details not described in detail in the embodiment can be referred to the above embodiment, and the embodiment has the same beneficial effects as the above embodiment.

[0157] Embodiment 3

[0158] On the basis of the first two embodiments, the embodiment provides an electric vehicle cluster multi-set meal master-slave game orderly charging optimization method. In order to verify the beneficial effects of the present application, economic benefit calculation and simulation experiments are used for scientific demonstration.

[0159] Table 1: Effect table of different electric vehicle charging modes

[0160]

[0161] Comparative Example 1: static load limitation is used, and electric vehicle charging is not guided;

[0162] As can be seen from Table 1, because there is no guidance for electric vehicle charging, the electric vehicle charging time conflicts with the resident electricity, and further leads to the obvious increase of the load at the peak time, even reaches the load, and in the case of the proportional 1, the car charging may face charging restriction or higher cost at the peak time, the waiting time also increases, which leads to the obvious lack of satisfaction of the embodiment 1.

[0163] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced without departing from the spirit and scope of the present application, and they should be covered in the scope of the claims of the present application.

Claims

1. A method for optimizing ordered charging of electric vehicle clusters using a master-slave game theory approach, characterized in that: Includes the following steps, Collect charging information and fit the probability distribution of charging demand for different charging piles at different times based on a probability prediction method; Electric vehicle charging packages are set up based on the probability distribution of charging demand and the power purchase and sale price of the distribution network. Based on charging information, a charging cost model is constructed, and combined with electric vehicle charging packages, a method for selecting electric vehicle charging load packages is formed. Methods for fitting the probability distribution of charging demand at different charging stations during different time periods include: Based on the improved LightGBM model, a single-point prediction objective function is set, expressed mathematically as follows: In the formula, L represents the objective function of the LightGBM model. Let be the predicted value of the charging pile in the t-th time period. Compared with the true value x t The error relationship between them, Ω(f) k ) represents the regularization term of the LightGBM model, γ and γ are both penalty coefficients, A represents the number of nodes in the k-th tree of the LightGBM model, and ω a This represents the weight of the a-th node in the k-th tree of the LightGBM model; The objective function is derived by taking the derivative of the i-th iteration using the forward distribution algorithm, and the number of weights ω is then calculated. a The objective function is obtained by finding the minimum value of , and the specific formula is as follows: I a ={i|q(x i )=a} In the formula, L (i) Let g be the objective function after the derivative of the i-th iteration of the LightGBM model, I represent the total number of iterations of the LightGBM model, ω represent the weight of the k-th tree in the LightGBM model, and g i and h i Let f represent the coefficients of the derivative in the i-th iteration, and f and f are both penalty coefficients. i (x i ) represents the derivative of the actual values ​​of the charging pile in the LightGBM model for each time period in the i-th iteration, I a Let ω represent the set of nodes in each tree, and let ω represent the weight of the k-th tree in the LightGBM model after the i-th iteration. A nonparametric kernel density estimation model is used to establish a probability distribution model for the charging demand prediction error of charging piles in different time periods. The mathematical formula is as follows: In the formula, denoted as x, where x′ represents the predicted value of the charging pile for each time period; and x′ represents the predicted error value of the charging pile for each time period. Let σ represent the probability distribution model of the charging demand prediction error for each time period, where σ represents the free parameters of the probability distribution model.

2. The electric vehicle cluster multi-package master-slave game-theoretic ordered charging optimization method as described in claim 1, characterized in that: The charging information includes information on the road network around the community, the vehicle owner's behavior trajectory, and the power grid topology; The information on the road network surrounding the community includes road network nodes, road connectivity, road grade, real-time traffic flow, and congestion status. The vehicle owner's behavior trajectory includes the probability distribution of travel time, travel time on the road segment, stay time at the destination, departure time of the trip, user's battery status, and charging time; The power grid topology includes transformer capacity, parameters of different branches in the community, node locations, and node parameters.

3. The electric vehicle cluster multi-package master-slave game-theoretic ordered charging optimization method as described in claim 2, characterized in that: The electric vehicle charging packages include fixed electricity price packages, peak-valley electricity price packages, and tiered electricity price packages; The fixed electricity price package model features a fixed electricity price throughout the day with no limit on the amount of electricity charged. The peak-valley electricity pricing package model is designed to guide electric vehicle charging load by setting peak-valley electricity prices, thereby participating in peak shaving and valley filling of the power grid. The tiered electricity pricing package model sets the charging load of electric vehicles into N levels, and implements different power consumption standards and electricity price standards for each level of the tiered increasing electricity price, so as to guide the agent to charge the electric vehicle charging load reasonably.

4. The electric vehicle cluster multi-package master-slave game-theoretic ordered charging optimization method as described in claim 3, characterized in that: In the fixed-price electricity package model, the electricity sales revenue of the m-th charging pile is expressed as: In the formula, This represents the revenue of the charging station during time period t on day d, based on a fixed electricity price package. This represents the fixed electricity price for the charging station during time period t on day d. This represents the electricity sold by the charging station during time period t on day d. In the aforementioned peak-valley electricity pricing model, the electricity sales revenue of the m-th charging pile is represented as: In the formula, This represents the revenue of a charging station during time period t on day d, based on the peak-valley electricity pricing package model. T represents the established peak-valley and off-peak electricity prices. P T M and T V This is represented as the established peak-valley-normal time period. and This represents the settlement price for charging stations during peak, valley, and normal periods. In the tiered electricity pricing model, the electricity sales revenue of the m-th charging pile is represented as: In the formula, This represents the revenue from the tiered electricity pricing package during time period t on day d. This represents the tiered electricity price for the charging station during time period t. and These represent the electricity prices for charging stations under different electricity consumption standards. This represents the electricity sold by the charging station during time period t on day d, according to the tiered electricity pricing package. and This represents the boundary between different electricity standards.

5. The electric vehicle cluster multi-package master-slave game-theoretic ordered charging optimization method as described in claim 4, characterized in that: The charging cost model includes an electricity purchase cost model, an investment and construction cost model, and an operation and maintenance cost model. The electricity purchase cost model is expressed mathematically as follows: The investment and construction cost model is expressed by the mathematical formula as follows: The operation and maintenance cost model is expressed by the mathematical formula as follows: The charging cost model is used to calculate the economic benefit model of the charging pile, specifically expressed as follows: F up =I ev -C inv -C yw -C gs In the formula, F up Represented as the economic benefits of charging stations, I ev C represents the revenue from electricity sales for charging station packages. inv C represents the investment and construction cost of charging stations. yw C represents the operation and maintenance cost of the charging station. gs ψ represents the cost of electricity purchased from the power grid by the charging pile, M represents the total number of charging piles, and ψ E γ represents the investment and construction cost per unit capacity of the charging pile, and γ represents the discount rate for all charging piles. This represents the lifespan of the m-th charging pile. This represents the rated capacity of the m-th charging pile. This represents the unit capacity operation and maintenance cost of the m-th charging pile. This represents the power purchased by the charging pile from the distribution network during time period t on day d. This represents the electricity price sold by the charging pile on the distribution network during time period t on day d. The formula for calculating the charging cost of electric vehicles is expressed as follows: F down =C ev +C loss The charging cost includes the cost of the package charging and the cost of lost charges. The cost of the package charging is expressed as follows: The cost of loss is expressed as: In the formula, F down C represents the charging cost of an electric vehicle charging load. ev C represents the charging cost of an electric vehicle package. loss Let λ represent the loss cost of electric vehicles, k1, k2, and k3 are all 0-1 parameters, and satisfy k1+k2+k3=1, i represents the i-th charging package of the charging pile, and λ represents the unit charging power cost loss of the electric vehicle charging load.

6. The electric vehicle cluster multi-package master-slave game-theoretic ordered charging optimization method as described in claim 5, characterized in that: The method for selecting electric vehicle charging load packages includes: using a master-slave game model to characterize the power interaction and dynamic game between charging piles and electric vehicle charging loads, and obtaining the Nash equilibrium solution by nesting the CPLEX commercial solver with the Cuckoo optimization algorithm; The master-slave game model between charging piles and electric vehicle charging loads can be represented as follows: G={N;S 充电桩 ;S 充电负荷 ;F up ;F down } Where N represents the participants, and their set is N = {charging piles ∪ electric vehicle charging loads}; S 充电桩 The main charging pile strategy is represented by the following set: S 充电负荷 The set of electric vehicle charging load strategies is represented as follows: In the formula, G represents the game theory model; Finally, the pricing results for charging pile packages and the selection method for electric vehicle charging load packages are derived from the obtained Nash equilibrium solution.

7. A multi-package master-slave game-theoretic ordered charging optimization system for electric vehicle clusters, based on the multi-package master-slave game-theoretic ordered charging optimization method for electric vehicle clusters according to any one of claims 1 to 6, characterized in that: It includes a charging data acquisition module, a charging package design module, a cost model construction module, and an optimization decision-making module; The charging data acquisition module acquires real-time charging-related data, including the charging demand of electric vehicles, vehicle location information, power grid status, and electricity price information. The charging package design module designs a variety of charging packages for electric vehicle users to choose from, based on the power grid's purchase and sale prices and the probability distribution of charging demand. The cost model building module is used to estimate and analyze the cost of the entire charging system, including electricity purchase cost, infrastructure construction cost, and operation and maintenance cost; The optimization decision module employs a master-slave game model from game theory and a nested business solver using the Cuckoo Search algorithm to provide optimization decisions for charging pile pricing and charging load management.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the electric vehicle cluster multi-package master-slave game-theoretic ordered charging optimization method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the electric vehicle cluster multi-package master-slave game-theoretic ordered charging optimization method according to any one of claims 1 to 6.

Images