V2G charging and discharging scheduling method and system based on improved backbone particle swarm optimization

Abstract

The invention discloses a V2G charging and discharging scheduling method and system based on an improved backbone particle swarm algorithm, and the method comprises the steps: estimating the load of an electric vehicle, and obtaining estimation data; constructing a charging and discharging scheduling model based on the estimation data, the power grid side demand and a preset target constraint; and introducing a learning factor, and solving the charging and discharging scheduling model based on an improved backbone particle swarm algorithm to obtain a scheduling strategy. The system comprises a statistical estimation module, a model construction module and a model solving module. By using the invention, a charging and discharging strategy of the electric vehicle is provided, the electric vehicle can be ensured to access the power grid orderly, and the use efficiency and stability of the power grid are improved. The method can be widely applied to the field of resource scheduling.

Description

Technical Field

[0001] The present invention relates to the field of resource scheduling, and in particular to a V2G charging and discharging scheduling method and system based on an improved backbone particle swarm algorithm. Background Art

[0002] With the development of renewable energy and autonomous driving technology, electric vehicles (EVs) have attracted widespread attention as an emerging mode of transportation. Electric vehicles have become increasingly popular due to their clean and efficient operation, low noise, high energy conversion efficiency and comfortable riding experience in urban environments, effectively addressing the current shortage of non-renewable energy and environmental pollution. However, the large-scale integration of electric vehicles into the power grid may significantly increase demand and pose challenges to the operational integrity of the existing power infrastructure. The randomness of electric vehicles connecting and disconnecting from the power grid may increase the instability of the power grid, affect its operation, and have a negative impact on the stability and reliability of the power grid. Therefore, it is necessary to design an electric vehicle charging and discharging scheduling strategy to improve the stability of power grid operation and resource utilization efficiency.

[0003] For this type of electric vehicle charging and discharging optimization scheduling mathematical model, the traditional optimization method relies on solving the initial point. Although the meta-heuristic method can achieve the optimal solution through rule selection and updating within a limited time, its algorithm efficiency and performance cannot meet the application scenario of this charging and discharging scheduling. Summary of the invention

[0004] In view of this, in order to solve the technical problems of low efficiency and poor performance of the metaheuristic method used in the existing charging and discharging scheduling method, the present invention proposes a V2G charging and discharging scheduling method based on an improved backbone particle swarm algorithm, and the method comprises the following steps:

[0005] Estimation of electric vehicle load based on Monte Carlo algorithm: The daily charging load of each electric vehicle is calculated using Monte Carlo algorithm, and then the daily charging load of each vehicle is accumulated to calculate the total charging load;

[0006] Construct a charging and discharging scheduling model based on estimated data, grid-side demand and preset target constraints;

[0007] A learning factor is introduced, and the charging and discharging scheduling model is solved based on an improved backbone particle swarm algorithm to obtain a scheduling strategy.

[0008] In some embodiments, the step of constructing a charging and discharging scheduling model based on estimated data, grid-side demand and preset target constraints specifically includes:

[0009] The V2G-based time-of-use electricity price strategy guides the charging time of electric vehicles, taking into account the economic needs of both the user side and the grid side, and establishes an objective function through three optimization indicators: grid peak-to-valley difference, grid fluctuation, and user charging cost;

[0010] A charging and discharging scheduling model is constructed based on preset constraints.

[0011] In some embodiments, the step of introducing a learning factor and solving the charging and discharging scheduling model based on an improved backbone particle swarm algorithm to obtain a scheduling strategy specifically includes:

[0012] Introduce learning factors and construct an improved backbone particle swarm algorithm;

[0013] The charging and discharging scheduling model is solved based on the improved backbone particle swarm algorithm;

[0014] In the solution process, the upper layer loop uses the Bayesian optimization strategy to find the optimal parameter set θ under the maximum number of iterations. best After that, it is used as the fixed parameter of the lower layer improved backbone particle swarm algorithm for calculation until the convergence condition of the improved backbone particle swarm algorithm is met, thereby obtaining the optimal solution of the model.

[0015] The present invention also proposes a V2G charging and discharging scheduling system based on an improved backbone particle swarm algorithm, the system comprising:

[0016] A statistical estimation module is used to estimate the load of electric vehicles and obtain estimated data;

[0017] A model building module, which builds a charging and discharging scheduling model based on the estimated data, grid-side demand and preset target constraints;

[0018] The model solving module introduces a learning factor and solves the charging and discharging scheduling model based on an improved backbone particle swarm algorithm to obtain a scheduling strategy.

[0019] Based on the above scheme, the present invention provides a V2G charging and discharging scheduling method and system of an improved backbone particle swarm algorithm, and proposes a charging and discharging strategy for electric vehicle-grid interaction that takes into account battery health and grid and user-side benefits as constraints; further, an improved intelligent optimization algorithm is proposed to improve the speed and accuracy of the solution, ensuring that an approximate optimal solution is found within a limited time; finally, a parameter optimization framework is designed by using a Bayesian optimization algorithm, the hyperparameters of the intelligent optimization algorithm are adaptively adjusted, the parameter setting process is simplified, and the optimization performance of the algorithm is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flowchart of the steps of a V2G charging and discharging scheduling method of the present invention using an improved backbone particle swarm algorithm;

[0021] Figure 2 is a load schematic diagram under the disordered charging strategy of a specific embodiment of the present invention;

[0022] Figure 3 is a schematic diagram of the Bayesian optimization convergence process of a specific embodiment of the present invention;

[0023] Figure 4 It is a schematic diagram of the load required by the algorithm under different initial parameters within the V2G strategy;

[0024] Figure 5 It is a structural block diagram of the hourly charge and discharge load of all electric vehicles of the present invention;

[0025] Figure 6 It is a structural block diagram of a V2G charging and discharging scheduling system of an improved backbone particle swarm algorithm of the present invention. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0027] It should be noted that, for the convenience of description, only the parts related to the relevant invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other.

[0028] It should be understood that the "system", "device", "unit" and / or "module" used in this application is a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the word can be replaced by other expressions.

[0029] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "a kind" and / or "the" do not refer to the singular, but also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of clearly identified steps and elements, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements. The elements defined by the sentence "includes a..." do not exclude the existence of other identical elements in the process, method, commodity or device that includes the elements.

[0030] In the description of the embodiments of the present application, "plurality" means two or more than two. The following terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features.

[0031] In addition, flow charts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, the various steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0032] Reference Figure 1 , which is a flow chart of an optional example of a V2G charging and discharging scheduling method of an improved backbone particle swarm algorithm proposed in the present invention. The method can be applied to computer equipment. The scheduling method proposed in this embodiment may include but is not limited to the following steps:

[0033] Step S1, estimating the load of the electric vehicle to obtain estimated data;

[0034] Step S2: constructing a charge and discharge scheduling model based on the estimated data, grid-side demand and preset target constraints;

[0035] Step S3: introducing a learning factor, solving the charging and discharging scheduling model based on an improved backbone particle swarm algorithm, and obtaining a scheduling strategy.

[0036] Each particle in the particle swarm is an N×24 matrix, where N is the number of vehicles connected to the grid in the problem, 24 is 24 hours a day, and any point in the matrix, such as (i, 1), represents the power compliance of the i-th vehicle at point 1. Furthermore, the number of particle swarms is initialized to an arbitrary tree. If the number of particle swarms is 100 at this time, there are 100 N×24 matrices for optimization and solution.

[0037] The purpose of the present invention is to mitigate grid load fluctuations by optimizing the charging and discharging strategies of electric vehicles, thereby smoothing the grid load curve and reducing the risk of grid instability caused by a large number of electric vehicles charging at the same time.

[0038] In some embodiments, the step S1 specifically includes:

[0039] Electric vehicle load estimation based on the Monte Carlo method obtains the daily driving distance and charging start time of electric vehicles through statistical data. Usually, the daily mileage of electric vehicles follows a log-normal distribution, and the following formula gives the probability density function of the daily mileage:

[0040]

[0041] Among them, L is a random variable, μ L is the average daily mileage, and σ is the standard deviation.

[0042] The return time t of electric vehicles follows a normal distribution, that is, The probability density function of the starting charging time of the electric vehicle is given by the following formula:

[0043]

[0044] Among them, the expected value μ c =17.6; standard deviation σ c =3.4.

[0045] Since the charging behavior of a single electric vehicle is highly random, according to the law of large numbers, the charging behavior of large-scale electric vehicles must have certain regularity. Therefore, the Monte Carlo algorithm can be used to calculate the daily charging load of each electric vehicle, and then the daily charging load of each vehicle can be accumulated to calculate the total charging load.

[0046] In some feasible embodiments, the step S2 specifically includes:

[0047] The V2G-based time-of-use electricity price strategy guides the charging time of electric vehicles, while taking into account the economic needs of the user side and the grid side. The setting of electricity prices is generally based on the peak and valley periods of the grid load. When the grid load is high, the electricity price is generally higher, and when the load is low, the electricity price also decreases accordingly. Users will guide their charging time according to the setting of electricity prices and incentives, change their original way of using electricity, and thus indirectly change the load distribution of the grid, while promoting the stability of the power market and the reliable operation of the grid. Therefore, the objective function is established through the three optimization indicators of the peak-to-valley difference of the grid, the grid fluctuation and the user charging cost.

[0048] User charging and discharging costs:

[0049] For users, through the guidance of time-of-use electricity prices, they can sell part of their electricity to the grid to reduce charging costs. However, due to the need for battery replacement, the V2G strategy indirectly shortens the battery life and increases the total cost of electric vehicles. Therefore, when establishing a user benefit model, it is necessary to consider both the reduction in annual charging costs and the increase in battery loss costs. Therefore, the battery loss cost C is defined as v (i,t) is:

[0050]

[0051] Among them, C v(i, t) is the battery loss cost, α is the linear relationship coefficient between battery life and cycle number, M is the battery replacement cost of EVs; E max (i) represents the maximum capacity of the EV battery, Γ v (i, t) is the cyclic charge and discharge capacity of EV at time t:

[0052] Γ v (i,t)=max{0, (Soc(i,t-1)-Soc(i,t))E max (i)}

[0053]

[0054] Among them, p(i, t) is the rated charging power of the i-th vehicle, η is the charging efficiency of EV, SOC represents the state of charge, and the Soc of EVs charging and discharging depends on the state of charge at the previous moment. When EVs are charging, Soc(i, t-1)-Soc(i, t)<0, and EVs do not incur battery loss costs. When EVs are discharging, Soc(i, t-1)-Soc(i, t)>0, and battery loss costs are incurred.

[0055] The user cost of charging and discharging is defined as:

[0056]

[0057] Among them, p c (i, t), p d (i, t) respectively represent the charging and discharging power of the i-th electric vehicle at time t, C c (t) and C d (t) represents the price of charging and discharging when connected to the grid at different times t.

[0058] Grid load peak-valley difference:

[0059] F 2obj =[max(P(t))-min(P(t))]

[0060]

[0061] Among them, P base (t) represents the basic load of the power grid at time t, and p(i, t) is the additional load brought by the electric vehicle after it is connected to the power grid.

[0062] Grid load fluctuations:

[0063]

[0064]

[0065] Among them, P av It indicates the average load of electric vehicles after they are connected to the power grid.

[0066] Since the dimensions of the three performance indicators are different, the normalization coefficients of the three objective functions are set as follows:

[0067] F 1e =P base (t)C av (t)

[0068] F 2e =max(P base (t))-min(P base (t))

[0069]

[0070] The final objective function can be expressed as:

[0071]

[0072] Among them, C av represents the average electricity price, a, b, c are the weight coefficients that need to be adjusted, which affect the final optimization effect.

[0073] Electric vehicle charging and discharging power constraints:

[0074]

[0075] p c (i, t) and p d (i, t) are the maximum charging and discharging power of the i-th electric vehicle at time t.

[0076] Maximum total load constraint of the power grid:

[0077]

[0078] Among them, P max (t) represents the maximum load of the power grid at time t.

[0079] EV expected capacity constraints:

[0080]

[0081] E d (i) = δE max (i)

[0082] Among them, E0(i, t) represents the initial remaining capacity of the battery after the electric vehicle is connected to the grid, and E d (i) represents the expected battery capacity and δ represents the expected percentage.

[0083] Electric vehicle state of charge constraints:

[0084]

[0085] Too high or too low SOC is harmful to the health of electric vehicle batteries. Therefore, the state of charge of EVs is within the constraint range. min (i) represents the minimum state of charge of the EV, Soc max (i) represents the maximum state of charge of the EV; E(i, t) represents the battery capacity of the i-th electric vehicle at the current moment.

[0086] In some feasible embodiments, the step S3 specifically includes:

[0087] S3.1. An improved backbone particle swarm optimization (IBBPSO) is proposed to solve the problem model. In the basic BBPSO algorithm, xpbest i，j and xgbest i，j The contribution to updating particles is the same from the beginning to the end, which is not conducive to the balance between global and local search. The IBBPSO algorithm proposed in this invention adds two learning factors, C1 and C2, to the Gaussian sampling formula to enhance the balance between individual influence and global influence. In order to accelerate the convergence of the algorithm, each component has a certain probability of selecting the global component xgbest at the same position. i,j , the improved formula is as follows:

[0088]

[0089] Among them, xpbest i，j is the j-th dimension component of the i-th historical optimal individual; xgbest i，j is the j-th dimension component of the i-th global optimal individual; C1 and C2 represent two learning factors; U(0,1) is the average distribution from 0 to 1, and pp is the probability of generating a population using the Gaussian sampling formula; Represents a normal distribution.

[0090] The present invention adopts a new method to deal with the problem of particle crossing the boundary. The crossing particle will randomly select a boundary value close to itself within a certain disturbance range to increase the diversity of solutions.

[0091]

[0092] The purpose of introducing mutation operation in particle swarm optimization is usually to avoid the algorithm from falling into local optimality. Through the perturbation of mutation operator, a certain dimension of decision variable of particle is located in other search areas, thereby increasing the diversity of solutions and improving the probability of finding global optimal solution.

[0093] S3.2, in the swarm intelligence optimization algorithm, the population size, the number of iterations, and the weights in the fitness function are all considered as adjustable hyperparameters, which have a significant impact on the final optimization results. Traditional adjustment methods may be too time-consuming and ineffective, so in the present invention, Bayesian optimization is combined to perform efficient automatic parameter adjustment.

[0094] Set the population size, number of iterations, and weights in the fitness function as hyperparameter sets:

[0095] θ:=[a, b, c, numP, maxIter]

[0096] In the present invention, the Gaussian process is selected as the proxy model for Bayesian optimization. The Gaussian process is determined when the mean m(θ) and covariance k(θ, θ′) are uniquely determined. The function to be evaluated is set to J(θ), and the mean and covariance functions of the Gaussian process are usually described as:

[0097] m(θ)=E[J(θ)]

[0098] k(θ, θ′)=coV(J(θ), J(θ′))

[0099] After having the proxy function, randomly select the initial point to build the proxy model, then select the next evaluation point of the proxy model through the acquisition function, calculate the new objective function value, and update the proxy model until the maximum number of iterations is reached. The acquisition function can be selected as the expected improvement E, and the acquisition function is expressed as:

[0100]

[0101] Among them, Φ(·) represents the probability density function, φ(·) represents the cumulative distribution, and the optimal sampling point θ is determined by minimizing the expected value of the acquisition function * :

[0102] θ*=argmin θ EI(θ)

[0103] Furthermore, in order to save model solving time, a penalty cost f(t) for time is introduced cost Finally, the objective function form for Bayesian optimization to solve the optimal hyperparameters is:

[0104] J θ =Fitness+f(t)cost

[0105] Among them, Fitness is about the IBBPSO algorithm to solve F obj The obtained fitness value.

[0106] S3.3. In the process of solving the model, the upper layer cyclically uses the Bayesian optimization strategy to find the optimal parameter set under the maximum number of iterations, and then uses it as the fixed parameters of the lower layer IBBPSO for calculation until the convergence conditions of IBBPSO are met, thereby obtaining the optimal solution of the model.

[0107] Based on the above solution, the present invention also provides actual data particles.

[0108] In the present invention, some basic parameter values ​​of electric vehicles are preset. The parameter status of the electric vehicle is given in Table 1.

[0109] Table 1 Basic parameters of electric vehicles

[0110] parameter Numeric Total number of electric vehicles 100 Energy consumption per mileage (km) 0.14286kwh / h Charging mode power 5kw Discharge mode power 3kw Average charging efficiency 90 Average discharge efficiency 90 Soc maximum level 0.95 Soc minimum level 0.1 Battery life cycle coefficient (output diagram) 0.0157 Battery costs 35000 Maximum battery capacity 72 Expected maximum capacity ratio 0.75

[0111] Firstly, the Monte Carlo simulation was used to simulate the daily charging load of 200 electric vehicles in a disorderly charging mode. Figure 2 As shown:

[0112] Further combined with time-of-use electricity prices and V2G strategies, the IBBPSO algorithm is used to calculate the load of the power grid after optimized scheduling. After the electric vehicle is connected to the power grid, it is charged and discharged according to the expected battery capacity required for the day's mileage, thereby reducing the peak-to-valley difference and fluctuation of the power grid while meeting the minimum power demand, and maximizing user benefits. In order to further illustrate the superiority of the algorithm proposed in the present invention, the optimization results of algorithms such as QPSO, PSO, CSO and IBBPSO are selected for comparison. Table 2 shows the load optimization results of the power grid and the user's benefits under different algorithm optimizations.

[0113] Table 2 Comparison results of power grid load under different algorithm optimization

[0114]

[0115] In order to further improve the optimization efficiency and optimization results of the algorithm, Bayesian optimization is used to adjust the initialization parameters of IBBPSO, and the calculated optimal fitness value and algorithm optimization time are considered in the evaluation function. Next, initialize the IBBPSO parameters θ0: = [a0, b0, c0, numP0, maxIter0], set the maximum number of Bayesian optimization iterations to 50, and the iterative process is as follows Figure 3 shown.

[0116] Figure 4 The calculated grid load curve is given. Figure 5 The charging and discharging loads of all EVs per hour are given.

[0117] The experimental results show that after electric vehicles are connected to the power grid, most of the charging time is concentrated between 23:00 and 5:00, and they discharge to the power grid during peak periods, thereby reducing the load on the power grid. The stability of the power grid load can be reflected by the peak-to-valley difference rate and load utilization rate. The optimized valley load is increased to 1363.9kw, the peak load is reduced to 1711.8kw, the peak-to-valley difference rate is reduced to 0.2032, and the load utilization rate of the power grid is increased to 0.8233, which is 14.49% higher than that of disorderly charging, and the charging cost of users is reduced by 13.66. Therefore, it can be considered that the V2G strategy based on time-of-use electricity prices has a significant effect on reducing the overall load fluctuation of the power grid and improving economic benefits. Table 3 shows the comparison of the basic load and the indicators of the power grid load before and after optimization.

[0118] Table 3 Comparison of basic load and grid load before and after optimization

[0119]

[0120]

[0121] like Figure 6 As shown, a V2G charging and discharging scheduling system based on an improved backbone particle swarm algorithm includes:

[0122] A statistical estimation module is used to estimate the load of electric vehicles and obtain estimated data;

[0123] A model building module, which builds a charging and discharging scheduling model based on the estimated data, grid-side demand and preset target constraints;

[0124] The model solving module introduces a learning factor and solves the charging and discharging scheduling model based on an improved backbone particle swarm algorithm to obtain a scheduling strategy.

[0125] The contents of the above method embodiments are all applicable to the present system embodiments. The functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0126] A V2G charging and discharging scheduling device based on improved backbone particle swarm algorithm:

[0127] at least one processor;

[0128] at least one memory for storing at least one program;

[0129] When the at least one program is executed by the at least one processor, the at least one processor implements the V2G charging and discharging scheduling method based on the improved backbone particle swarm algorithm as described above.

[0130] The contents of the above method embodiments are all applicable to the present device embodiments. The functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0131] A storage medium stores processor-executable instructions, which are used to implement a V2G charging and discharging scheduling method based on an improved backbone particle swarm algorithm when executed by the processor.

[0132] The contents of the above method embodiments are all applicable to the present storage medium embodiments. The functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0133] The above is a specific description of the preferred implementation of the present invention, but the invention is not limited to the embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A V2G charging and discharging scheduling method based on an improved backbone particle swarm algorithm, characterized in that: The following steps are involved: Estimating the load of electric vehicles to obtain estimated data; Based on the estimated data, grid-side demand and preset target constraints, a charging and discharging scheduling model is constructed; A learning factor is introduced, and the charging and discharging scheduling model is solved based on an improved backbone particle swarm algorithm to obtain a scheduling strategy.

2. According to the V2G charging and discharging scheduling method of the improved backbone particle swarm algorithm of claim 1, it is characterized in that: The process of solving the model also includes: The improved backbone particle swarm algorithm is automatically tuned using a Bayesian optimization strategy to obtain optimal parameters.

3. According to the V2G charging and discharging scheduling method of the improved backbone particle swarm algorithm of claim 2, it is characterized in that: The step of estimating the electric vehicle load to obtain estimated data specifically includes: The electric vehicle load is estimated based on the Monte Carlo method, and the probability density function of daily mileage is obtained; Based on the probability density function of the daily mileage, the daily charging load and the total charging load of the electric vehicle are calculated.

4. According to the V2G charging and discharging scheduling method of the improved backbone particle swarm algorithm of claim 2, it is characterized in that: The step of constructing a charging and discharging scheduling model based on the estimated data, grid-side demand and preset target constraints specifically includes: With the goal of taking into account the economic needs of both the user side and the grid side, the objective function is established by taking the peak-to-valley difference of the grid, grid fluctuation and user charging cost as optimization indicators; Based on the objective function, a charging and discharging scheduling model is constructed in combination with the electric vehicle charging and discharging power constraints, the maximum total load constraints of the power grid, the electric vehicle expected capacity constraints and the electric vehicle charge state constraints.

5. According to the improved backbone particle swarm algorithm V2G charging and discharging scheduling method of claim 4, it is characterized in that: The objective function is expressed as follows: F 2obj =[max(P(t))-min(P(t))] Among them, F 1obj Indicates the user's charging and discharging cost; F 2obj Indicates the peak-to-valley difference of power grid load; F 3obj Indicates grid load fluctuation; F 1E 、F 2E 、F 3e Respectively represent different normalization coefficients; a, b and c represent different weight coefficients; p c (i,t) represents the maximum charging power of the i-th electric vehicle at time t; p d (i,t) represents the maximum discharge power of the i-th electric vehicle at time t; C c (t) and C d (t) represents the charging and discharging price of accessing the power grid at different times t; C v (i, t) is the battery loss cost; T is the maximum value at the time; N is the total number of electric vehicles; P(t) represents the total load of the power grid at time t; P base (t) represents the basic load of the power grid at time t; p(i,t) represents the additional load brought by the electric vehicle after it is connected to the power grid; P av It indicates the average load of electric vehicles after they are connected to the power grid.

6. The V2G charging and discharging scheduling method based on improved backbone particle swarm algorithm according to claim 5 is characterized by: The formula for the charging and discharging power constraint of electric vehicles is as follows: The formula for the maximum total load constraint of the power grid is as follows: The formula for the expected capacity constraint of electric vehicles is as follows: E d (i)=δE max (i) Among them, E0(i,t) represents the initial remaining capacity of the battery after the electric vehicle is connected to the grid, and E d (i) represents the expected battery capacity, δ represents the expected percentage; E max (i) indicates the maximum capacity of the battery; The formula for the state of charge constraint of electric vehicles is as follows: Among them, Soc min (i) represents the minimum state of charge of the battery, Soc max (i) represents the maximum state of charge of the battery, and E(i,t) represents the battery capacity of the i-th electric car at the current moment.

7. The V2G charging and discharging scheduling method based on improved backbone particle swarm algorithm according to claim 6 is characterized in that: The formula of the improved backbone particle swarm algorithm is as follows: Among them, xpbest i,j is the j-th dimension component of the i-th historical optimal individual; xgbest i,j is the j-th dimension component of the i-th global optimal individual; C1 and C2 represent two learning factors; U(0,1) is the average distribution from 0 to 1, and pp is the probability of generating a population using the Gaussian sampling formula; Represents a normal distribution.

8. A V2G charging and discharging scheduling system based on an improved backbone particle swarm algorithm, characterized in that: include: A statistical estimation module is used to estimate the load of electric vehicles and obtain estimated data; A model building module, which builds a charging and discharging scheduling model based on the estimated data, grid-side demand and preset target constraints; The model solving module introduces a learning factor and solves the charging and discharging scheduling model based on an improved backbone particle swarm algorithm to obtain a scheduling strategy.

9. A V2G charging and discharging scheduling device based on an improved backbone particle swarm algorithm, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the V2G charging and discharging scheduling method of the improved backbone particle swarm algorithm as described in any one of claims 1-7.

Images