Frequency fluctuation stabilizing method based on V2G technology

Abstract

The invention discloses a V2G technology-based frequency fluctuation stabilizing method. The method comprises the following steps of S1, constructing an objective function for measuring a power dispatching center to stabilize frequency fluctuation by using a V2G technology and constraint conditions when an electric vehicle participates in frequency modulation auxiliary service; s2, considering constraint conditions, and adopting a multi-objective genetic algorithm (MOGA) to solve the objective function to obtain an optimal solution set; and S3, the frequency modulation effect is judged according to the frequency deviation square root mean value of each solution in the optimal solution set, and finally the charging and discharging power of the electric vehicle is obtained. The influence of the electric vehicle participating in frequency modulation on stabilizing frequency fluctuation is improved, so that the power quality of a power grid is improved.

Description

Technical Field

[0001] The present invention relates to the field of new energy vehicles and electric power dispatching technology, and in particular to a method for suppressing frequency fluctuations by allowing electric vehicles to participate in frequency modulation based on V2G technology. Background Art

[0002] As the number of electric vehicles continues to increase, the disorderly charging of large-scale electric vehicles connected to the power grid will increase grid losses, reduce power quality, and increase the difficulty of optimizing and controlling grid operations. In the current situation where the problem of new energy consumption is serious, it poses a great threat to the safe and stable operation of the power grid. Therefore, it is necessary to optimize and guide the charging of electric vehicles.

[0003] At the same time, the high volatility and intermittent nature of renewable energy sources such as wind power, hydropower, and photovoltaics have put tremendous pressure on the frequency stability of the power system. Traditional units have low ramp rates and slow response speeds, so they are unable to timely adjust the impact of renewable energy uncertainty on the frequency disturbance of the power system. Electric vehicles, however, can be connected to the power grid as mobile energy storage systems and peak-shaving devices under the unified dispatch of the control center. Their functions of peak shaving and valley filling, frequency regulation, etc. can greatly improve the reliability and economy of power system operation.

[0004] However, how operators determine the charging and discharging time and charging and discharging power of electric vehicles in their jurisdiction to improve the load curve is a key issue that needs to be solved. Due to the large number of electric vehicles and the different available time, initial battery capacity, charging and discharging power limits and minimum battery capacity set by users for each electric vehicle, this problem is a high-dimensional, multi-variable, and multi-constrained complex optimization problem, which urgently requires an optimization algorithm that is less affected by the dimension of the optimization problem to solve it. Summary of the invention

[0005] The purpose of the present invention is to solve the deficiencies of the above-mentioned prior art, thereby providing a frequency modulation method for smoothing frequency fluctuations based on V2G technology, improving the influence of electric vehicles participating in frequency modulation on smoothing frequency fluctuations, and thus improving the power quality of the power grid.

[0006] A method for smoothing frequency fluctuations based on V2G technology comprises the following steps:

[0007] S1: Construct an objective function to measure the use of V2G technology to smooth frequency fluctuations by the power dispatch center and the constraints when electric vehicles participate in frequency regulation auxiliary services;

[0008] S2: Considering the constraints, a multi-objective genetic algorithm (MOGA) is used to solve the objective function and obtain the optimal solution set;

[0009] S3: The frequency modulation effect is judged by the root mean square of the frequency deviation of each solution in the optimal solution set, and finally the charging and discharging power of the electric vehicle is obtained.

[0010] The objective function for measuring the use of V2G technology to smooth frequency fluctuations in the power dispatching center is as follows:

[0011] The following objective function is established based on the optimal combination of the two objectives of minimizing the frequency regulation cost under the participation of electric vehicles and achieving the best tracking effect of the frequency regulation power demand signal:

[0012]

[0013] Where ψ represents the frequency error. The closer it is to 0, the better the tracking effect. 1,i (t) represents the frequency regulation cost of electric vehicle i at time t.

[0014] The expression of frequency error ψ is as follows:

[0015]

[0016] w=T total / Δt,T total is the total frequency modulation duration, Δt is the sampling time interval, N is the number of vehicles participating in the frequency modulation, the subscript i represents the vehicle number, P EVi,k is the charging and discharging power of the i-th vehicle in the k-th sampling period, is the frequency modulation power required by the power grid in the kth sampling period.

[0017] The cost function of frequency regulation cost with the participation of electric vehicles is as follows:

[0018]

[0019]

[0020]

[0021]

[0022] In the formula, I is the capacity investment cost, energy loss cost, and life loss cost of electric vehicle i; f，t It is an indicator used to calculate the compound interest effect; P AGC Indicates the power output required by automatic power generation control; s e 、c P,i is the unit capacity cost, grid-connected electricity price, and unit power cost of electric vehicle i; is the charging power, discharging power and rated power of electric vehicle i at time t; is the charging and discharging efficiency of the energy storage battery of electric vehicle i; is the rated capacity of electric vehicle i; r is the annual interest rate, which is 8%; T f,i is the floating charge life of the energy storage battery of electric vehicle i; N o,i is the number of cycles of the energy storage battery of electric vehicle i under full charge and discharge conditions, t oc is the scheduling times; Δt is the optimization period.

[0023] The constraints for electric vehicles to participate in frequency regulation auxiliary services are:

[0024] Constraints on charge and discharge current:

[0025]

[0026] Line power capacity constraints:

[0027]

[0028]

[0029]

[0030]

[0031] In the formula, I ic ——is the charging current value of electric vehicle i;

[0032] I id ——is the discharge current value of electric vehicle i;

[0033] I iN ——It depends on the battery model of the electric vehicle. The corresponding value for an 80Ah battery is 80A;

[0034] V i,t ——Real-time voltage value of the energy storage battery;

[0035] ——The lower limit of charging (discharging) power of electric vehicle i in time period j;

[0036] ——The upper limit of charging (discharging) power of electric vehicle i in time period j;

[0037] The constraint conditions and upper and lower limits calculation methods of electric vehicle rated capacity are as follows:

[0038]

[0039]

[0040]

[0041]

[0042] Wherein, SOC——is the state of charge of the battery, that is, the ratio of the remaining capacity of the battery to the maximum capacity;

[0043] SOC i,t ——is the real-time capacity state of electric vehicle i at time t;

[0044] SOC min ——is the lowest value of SOC considering the constraints of the battery itself, taking 0.2;

[0045] SOC max ——is the maximum value of SOC, taking 1;

[0046] One is the rated capacity value of the electric vehicle, in unit of Ah;

[0047] One is the remaining capacity of the battery of electric vehicle i in the time period t, in unit of Ah;

[0048] ΔE B,it One is the charge (discharge) capacity of electric vehicle i in the time period t;

[0049] One is the lower limit of the available capacity of electric vehicle i in the time period t;

[0050] One is the upper limit of the available capacity of electric vehicle i in the time period t;

[0051] Available time constraint of the vehicle: When the electric vehicle is not connected to the network, its available power is zero;

[0052] Minimum battery capacity constraint set by the user when the vehicle leaves:

[0053] SOC ik ≥SOC i,set

[0054] Wherein, SOC i,set ——is the lowest SOC (set by the user) when vehicle i leaves the station;

[0055] SOC ik ——is the SOC of vehicle i when it leaves the station;

[0056] K——is the departure time of the electric vehicle from the station.

[0057] Step S2 is specifically as follows:

[0058] S21: Initialize the population: Randomly generate a set of initial solutions, called the population, and each solution is usually represented by a set of variables or parameters called genes;

[0059] S22: Fitness evaluation: For each individual (solution), the established objective function is used to evaluate its performance on multiple objectives, and the degree of individual adaptation to multiple objectives is also quantified;

[0060] S23: Non-dominated sorting: According to the value of the fitness function, the individuals in the population are sorted non-dominatedly. Non-dominated sorting divides them into different levels according to the dominance relationship between individuals, so that individuals in each level are not dominated by individuals in other levels;

[0061] S24: Crowding calculation: For each individual in each level, calculate its crowding. Crowding is a measure of the distribution density of individuals in the solution space, which is used to maintain the diversity of the population.

[0062] S25: Selection operation: According to the non-dominated sorting and crowding calculation, the roulette operation is used to select the next generation of individuals;

[0063] S26: Update population: Update the current population to form the next generation population based on selection, crossover and mutation operations;

[0064] S27: Determine whether the dispatching target of the dispatching center is met;

[0065] S28: When the termination condition is met, that is, when the scheduling cost and frequency deviation are less than the set value (selected according to the requirements of the scheduling center), the high-level individuals in the non-dominated sorting are returned as the output of the algorithm. These individuals constitute a set of optimal solutions to the problem, namely the Pareto solution set.

[0066] The expression of the square root mean of frequency deviation is:

[0067]

[0068] In the formula, △f i Represents the i-th frequency deviation value, the root mean square value of the frequency deviation f rms The smaller it is, the smaller the frequency fluctuation is and the better the FM effect is.

[0069] After obtaining and classifying the travel parameters of electric vehicles, the present invention analyzes the goals that the power grid dispatching center needs to achieve for the electric vehicles to participate in frequency modulation, establishes a multi-objective V2G optimization dispatching model, solves the model through a multi-objective genetic algorithm (MOGA), and after obtaining the optimal Pareto solution set, analyzes the priority between multiple objectives to obtain the optimal compromise solution, thereby improving the impact of electric vehicles participating in frequency modulation on smoothing frequency fluctuations, thereby improving the power quality of the power grid.

[0070] The present invention is based on a multi-objective genetic algorithm (MOGA), which constructs a multi-objective function with the lowest frequency modulation cost and the best frequency tracking effect for solving, and determines the frequency modulation power borne by the V2G station composed of each electric vehicle, thereby reducing the overall frequency modulation cost and improving the sustainability of frequency modulation.

[0071] In the present invention, the input variables of the multi-objective genetic algorithm are the real-time measurement data obtained by the dispatching center, including the system frequency, the interconnection line switching frequency, etc., and the electric vehicle parameters identified by the electric vehicle control center, and the output variables are the minimum frequency regulation cost function and the minimum tracking effect function ψ. Compared with the prior art, it is more practical to consider the constraints between the frequency regulation cost and the frequency regulation effect when controlling the charging and discharging power of electric vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 This is the V2G frequency modulation control strategy diagram;

[0073] Figure 2 is a flow chart of the method of the present invention;

[0074] Figure 3 Flowchart for solving the model based on MOGA algorithm. DETAILED DESCRIPTION

[0075] The present invention is described in detail below in conjunction with the accompanying drawings. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process:

[0076] The grid control center can treat the electric vehicle capacity under the jurisdiction of an electric vehicle control center as a special power source. This power source is different from the traditional power source. The response power range of the traditional power source is a range greater than zero, while the response range of V2G can extend to negative values. Based on the adjustable capacity range and load forecast information provided by the electric vehicle control center, the grid control center sends a dispatch command to the electric vehicle control center to adjust the frequency.

[0077] The electric vehicle control center manages and controls all electric vehicles, allowing all the electric vehicles it manages to participate in power system services as a whole.

[0078] The battery owner can decide whether to participate in grid services when the battery is idle. The user can send a signal to inform the electric vehicle control center and communicate in real time to upload electric vehicle information. Electric vehicle owners can log in to the Web user interface and complete settings, updates, and modifications according to travel needs, thereby updating database information. In addition, the status of electric vehicles can be queried, and real-time electricity price information can be queried at any time when the power market mechanism matures.

[0079] The electric vehicle control center needs to query its database to obtain information about each electric vehicle, including: access location, available time, minimum battery capacity set by the user, and real-time battery capacity status. By monitoring the status of each electric vehicle, the overall available capacity at any time is predicted and the results are uploaded to the power grid control center.

[0080] The grid control center determines the total power instructions (charging and discharging plans) of each electric vehicle control center based on the load forecast (peak shaving service demand, frequency regulation service demand, peak shaving service revenue unit price, frequency regulation service revenue unit price and time-of-use electricity purchase price) and V2G available capacity forecast information.

[0081] For a large number of electric vehicles, a decentralized access and centralized control management model is adopted. The entire control process is a hierarchical control model, including two control layers: the power grid control center sends dispatch commands to the electric vehicle control center and the electric vehicle control center distributes power to individual electric vehicles. The realization of two-way energy exchange between electric vehicles and the power grid involves three independent entities: the power grid control center, the electric vehicle control center and the electric vehicle.

[0082] V2G frequency modulation control strategy:

[0083] like Figure 1 This is the principle diagram of optimizing and controlling the grid frequency using V2G technology proposed in the present invention. The principle framework mainly includes four modules: judgment module, frequency modulation power calculation module, MOGA control module and electric vehicle intelligent charging module. The frequency modulation power calculation module converts the load frequency control signal transmitted by the grid dispatching center into a frequency modulation power demand signal and transmits it to the MOGA control module. Based on the frequency modulation power demand signal and the SOC information collected by the electric vehicle control center, the module optimizes the charging and discharging power of each electric vehicle participating in V2G frequency modulation with the minimum frequency modulation cost and the best frequency modulation power demand signal tracking effect as the optimization goal, and then sends the power control signal to the electric vehicle intelligent charging module.

[0084] like Figure 2 As shown, the method for smoothing frequency fluctuations based on V2G technology includes the following steps:

[0085] S1: Construct an objective function to measure the use of V2G technology to smooth frequency fluctuations by the power dispatch center and the constraints when electric vehicles participate in frequency regulation auxiliary services;

[0086] The following objective function is established based on the optimal combination of the two objectives of minimizing the frequency regulation cost under the participation of electric vehicles and achieving the best tracking effect of the frequency regulation power demand signal:

[0087]

[0088] Where ψ represents the frequency error. The closer it is to 0, the better the tracking effect. 1,i (t) represents the frequency regulation cost of electric vehicle i at time t.

[0089] The controller's tracking effect on the frequency modulation power demand signal is evaluated based on the error ψ, which is expressed as follows:

[0090]

[0091] Where w = T total / Δt,T total is the total frequency modulation duration, Δt is the sampling time interval, and the closer ψ is to 0, the better the surface tracking effect. N is the number of vehicles participating in the frequency modulation, the subscript i represents the vehicle number, and P EVi,k is the charging and discharging power of the i-th vehicle in the k-th sampling period, is the frequency modulation power required by the power grid in the kth sampling period.

[0092] When electric vehicles participate in grid frequency regulation services, their capacity investment costs and energy loss costs must be considered. In addition, electric vehicles need to frequently change their output size and direction during frequency regulation, which will cause the life of their energy storage batteries to decay. Therefore, the cost function modeling of electric vehicles participating in frequency regulation must consider their life depreciation costs. Therefore, their frequency regulation costs are as follows:

[0093]

[0094]

[0095]

[0096] In the formula, is the capacity investment cost, energy loss cost, and life depreciation cost of electric vehicle i; s e 、c P,i is the unit capacity cost, grid-connected electricity price, and unit power cost of electric vehicle i; is the charging power, discharging power and rated power of electric vehicle i at time t; is the charging and discharging efficiency of the energy storage battery of electric vehicle i; is the rated capacity of electric vehicle i; r is the annual interest rate, which is 8%; T f,i is the floating charge life of the energy storage battery of electric vehicle i; N o,i is the number of cycles of the energy storage battery of electric vehicle i under full charge and discharge conditions, t oc is the scheduling times; Δt is the optimization period.

[0097] Therefore, the cost function of electric vehicles participating in frequency regulation is as follows:

[0098]

[0099] In the formula, F 1,i (t) represents the frequency regulation cost of electric vehicle i at time t

[0100] The power constraints in the multi-objective genetic algorithm are as follows:

[0101] The power constraint is mainly reflected in the constraint of charge and discharge current. The maximum value of charge current is 1 / 3I iN , the maximum discharge current is 2I iN .

[0102]

[0103] Considering the line power capacity constraint, it should not exceed 15kW;

[0104]

[0105]

[0106] Combining the above formulas, we get:

[0107]

[0108]

[0109] Where I ic ——is the charging current value of electric vehicle i;

[0110] I id ——is the discharge current value of electric vehicle i;

[0111] I iN ——It depends on the battery model of the electric vehicle. The corresponding value for an 80Ah battery is 80A;

[0112] V i,t ——Real-time voltage value of the energy storage battery;

[0113] ——The lower limit of charging (discharging) power of electric vehicle i in time period j;

[0114] ——The upper limit of charging (discharging) power of electric vehicle i in time period j.

[0115] The capacity of electric vehicle batteries is certain, the rated capacity It refers to the discharge capacity of the battery from a fully charged state to the discharge termination voltage under certain discharge conditions. The capacity of the battery cannot exceed the rated capacity value, and the battery life capacity should not be less than 20% of the battery capacity.

[0116]

[0117]

[0118]

[0119]

[0120] where SOC is the state of charge of the battery, i.e., the ratio of the remaining capacity of the battery to the maximum capacity;

[0121] SOC i,t —— is the real-time capacity state of electric vehicle i at time t;

[0122] SOC min —— is the minimum value of SOC considering the constraints of the battery itself, taking 0.2;

[0123] SOC max —— is the maximum value of SOC, taking 1.

[0124] One is the rated capacity value of the electric vehicle, in Ah;

[0125] One is the remaining capacity of the battery of electric vehicle i in the t period, in Ah;

[0126] ΔE B,it One is the charge (discharge) capacity of electric vehicle i in the period t;

[0127] —— is the lower limit of the available capacity of electric vehicle i in the period t;

[0128] —— is the upper limit of the available capacity of electric vehicle i in the period t.

[0129] Since the most basic function of the vehicle is still a means of transportation, it is impossible for an electric vehicle to be connected to the network every hour. To be more in line with the actual situation, the available time constraint of the vehicle is considered in this paper. When the electric vehicle is not connected to the network, its available power is zero.

[0130] Each user can set the minimum battery capacity when the vehicle leaves, considering the electricity demand for the user's trip. There are also the following constraint conditions:

[0131] SOC ik ≥SOC i,set

[0132] where SOC i,set —— is the minimum SOC (set by the user) when vehicle i leaves the station;

[0133] K——The time when the electric car leaves the station.

[0134] S2: Considering the constraints, a multi-objective genetic algorithm (MOGA) is used to solve the objective function and obtain the optimal solution set;

[0135] The optimization process of the target genetic algorithm is as follows Figure 3 As shown, the specific steps are:

[0136] S21: Initialize the population: Randomly generate a set of initial solutions, called the population, each solution is usually represented by a set of variables or parameters called genes;

[0137] S22: Fitness evaluation: For each individual (solution), the established objective function is used to evaluate its performance on multiple objectives, and the degree of individual adaptation to multiple objectives is also quantified;

[0138] S23: Non-dominated sorting: According to the value of the fitness function, the individuals in the population are sorted non-dominatedly. Non-dominated sorting divides them into different levels according to the dominance relationship between individuals, so that individuals in each level are not dominated by individuals in other levels;

[0139] S24: Crowding calculation: For each individual in each level, calculate its crowding. Crowding is a measure of the distribution density of individuals in the solution space, which is used to maintain the diversity of the population.

[0140] S25: Selection operation: According to the non-dominated sorting and crowding calculation, the roulette operation is used to select the next generation of individuals;

[0141] S26: Update population: Update the current population to form the next generation population based on selection, crossover and mutation operations;

[0142] S27: Determine whether the dispatching target of the dispatching center is met;

[0143] S28: When the termination condition is met, that is, when the scheduling cost and frequency deviation are less than the set value (selected according to the requirements of the scheduling center), the high-level individuals in the non-dominated sorting are returned as the output of the algorithm. These individuals constitute a set of optimal solutions to the problem, namely the Pareto solution set.

[0144] S3: The frequency modulation effect is judged by the root mean square of the frequency deviation of each solution in the optimal solution set, and finally the charging and discharging power of the electric vehicle is obtained.

[0145] The conflicts between objectives should be considered first in multi-objective problems. The optimal value of a single objective may affect other constraints and cannot be accepted by other objective values. Multi-objective optimization obtains multiple sets of Patero optimal solutions, and the most suitable one needs to be found from these optimal solutions. The frequency modulation effects of different solutions can be judged based on the square root mean of the frequency deviation:

[0146]

[0147] The root mean square value of the system frequency deviation f rms The smaller it is, the smaller the frequency fluctuation is and the better the strategy frequency modulation effect is.

[0148] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for smoothing frequency fluctuations based on V2G technology, characterized in that , including the following steps: S1: Construct an objective function to measure the use of V2G technology to smooth frequency fluctuations by the power dispatch center and the constraints when electric vehicles participate in frequency regulation auxiliary services; S2: Considering the constraints, a multi-objective genetic algorithm (MOGA) is used to solve the objective function and obtain the optimal solution set; S3: The frequency modulation effect is judged by the root mean square of the frequency deviation of each solution in the optimal solution set, and finally the charging and discharging power of the electric vehicle is obtained.

2. The frequency modulation method for smoothing frequency fluctuations based on V2G technology according to claim 1, characterized in that: The objective function for measuring the use of V2G technology to smooth frequency fluctuations in the power dispatching center is as follows: The following objective function is established based on the optimal combination of the two objectives of minimizing the frequency regulation cost under the participation of electric vehicles and achieving the best tracking effect of the frequency regulation power demand signal: Where ψ represents the frequency error. The closer it is to 0, the better the tracking effect. 1,i (t) represents the frequency regulation cost of electric vehicle i at time t.

3. The frequency modulation method for smoothing frequency fluctuations based on V2G technology according to claim 2, characterized in that: The expression of ψ is as follows: w=T total / Δt,T total is the total frequency modulation duration, Δt is the sampling time interval, N is the number of vehicles participating in the frequency modulation, the subscript i represents the vehicle number, P EVi,k is the charging and discharging power of the i-th vehicle in the k-th sampling period, P optk is the frequency modulation power required by the power grid in the kth sampling period.

4. The frequency modulation method for smoothing frequency fluctuation based on V2G technology according to claim 1, 2 or 3, characterized in that: The cost function of frequency regulation cost with the participation of electric vehicles is as follows: In the formula, I is the capacity investment cost, energy loss cost, and life loss cost of electric vehicle i; f，t It is an indicator used to calculate the compound interest effect; P AGC Indicates the power output required by automatic power generation control; s e 、c P,i is the unit capacity cost, grid-connected electricity price, and unit power cost of electric vehicle i; is the charging power, discharging power and rated power of electric vehicle i at time t; is the charging and discharging efficiency of the energy storage battery of electric vehicle i; is the rated capacity of electric vehicle i; r is the annual interest rate, which is 8%; T f,i is the floating charge life of the energy storage battery of electric vehicle i; N o,i is the number of cycles of the energy storage battery of electric vehicle i under full charge and discharge conditions, t oc is the scheduling times; Δt is the optimization period.

5. The frequency modulation method for smoothing frequency fluctuation based on V2G technology according to claim 1, characterized in that: The constraints for electric vehicles to participate in frequency regulation auxiliary services are: Constraints on charge and discharge current: Line power capacity constraints: In the formula, I ic ——is the charging current value of electric vehicle i; I id ——is the discharge current value of electric vehicle i; I iN ——It depends on the battery model of the electric vehicle. The corresponding value for an 80Ah battery is 80A; V i,t ——Real-time voltage value of the energy storage battery; ——The lower limit of charging (discharging) power of electric vehicle i in time period j; ——The upper limit of charging (discharging) power of electric vehicle i in time period j; The constraint conditions and upper and lower limits calculation methods of electric vehicle rated capacity are as follows: In the formula, SOC is the battery power state, that is, the ratio of the remaining battery capacity to the maximum capacity; SOC i,t ——is the real-time capacity status of electric vehicle i at time t; SOC min ——The minimum SOC value when considering the battery's own constraints is 0.2; SOC max —— is the maximum value of SOC, which is 1; One is the rated capacity of the electric vehicle, in Ah; 1 is the remaining battery capacity of electric vehicle i at time period t, in Ah; ΔE B,it One is the charging (discharging) capacity of electric vehicle i in time period t; is the lower limit of the available capacity of the electric vehicle i at time period t; - The upper limit of the available capacity of the electric vehicle i at time period t; Vehicle availability time constraint: When an electric vehicle is not connected to the grid, its available power is zero; The user sets the minimum battery capacity constraint when the vehicle leaves: SOC ik ≥SOC i,set In the formula, SOC i,set ——Minimum SOC of vehicle i when leaving the station (user set); SOC ik ——SOC of vehicle i when it goes into battle; K——The time when the electric car leaves the station.

6. The frequency modulation method for smoothing frequency fluctuation based on V2G technology according to claim 1, characterized in that: Step S2 is specifically as follows: S21: Initialize the population: Randomly generate a set of initial solutions, called the population, each solution is usually represented by a set of variables or parameters called genes; S22: Fitness evaluation: For each individual (solution), the established objective function is used to evaluate its performance on multiple objectives, and the degree of individual adaptation to multiple objectives is also quantified; S23: Non-dominated sorting: According to the value of the fitness function, the individuals in the population are sorted non-dominatedly. Non-dominated sorting divides them into different levels according to the dominance relationship between individuals, so that individuals in each level are not dominated by individuals in other levels; S24: Crowding calculation: For each individual in each level, calculate its crowding. Crowding is a measure of the distribution density of individuals in the solution space, which is used to maintain the diversity of the population. S25: Selection operation: According to the non-dominated sorting and crowding calculation, the roulette operation is used to select the next generation of individuals; S26: Update population: Update the current population to form the next generation population based on selection, crossover and mutation operations; S27: Determine whether the dispatching target of the dispatching center is met; S28: When the termination condition is met, the high-level individuals in the non-dominated sorting are returned as the output of the algorithm. These individuals constitute a set of optimal solutions to the problem, namely the Pareto solution set.

7. The frequency modulation method for smoothing frequency fluctuation based on V2G technology according to claim 1, characterized in that: The expression of the square root mean of frequency deviation is: In the formula, △f i Represents the i-th frequency deviation value, the root mean square value of the frequency deviation f rms The smaller it is, the smaller the frequency fluctuation is and the better the FM effect is.

Images