A method for arranging battery swap stations considering the demand for battery swapping of electric vehicles and the carrying capacity of the power grid

Through Monte Carlo simulation and particle swarm algorithm, the charging load prediction and battery swap station layout of electric vehicle are optimized, and the unreasonable layout of electric vehicle charging and swap stations is solved, and the optimization and stability of distribution network load is achieved.

CN118713077BActive Publication Date: 2024-11-26STATE GRID ANHUI ELECTRIC POWER CO LTD ELECTRIC POWER SCI RES INST +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411208179.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-30
Publication Date
2024-11-26
Estimated Expiration
2044-08-30

AI Technical Summary

Technical Problem

The strong randomness of electric vehicle charging and swapping stations and complex constraints lead to unreasonable layout of battery swapping stations, resulting in high risk of overloading and operation of distribution networks.

Method used

Monte Carlo simulation method is used to predict the charging load of electric vehicles, and a multi-objective model is established to target the minimum capacity-load ratio of the distribution network transformer and line, the minimum distribution network loss and the minimum voltage offset. The particle swarm algorithm is used to solve the problem. Optimal planning of the layout of battery swap stations.

Benefits of technology

By comprehensively considering the random load characteristics of electric vehicles, optimizing the layout of the battery swap station, comprehensively and accurately reflecting the power grid status, reducing the distribution network load, and improving the grid stability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118713077B_ABST
    Figure CN118713077B_ABST
Patent Text Reader

Abstract

A method for arranging battery swap stations taking into account the battery swapping demand of electric vehicles and the carrying capacity of power grids belongs to the field of optimized layout of battery swap stations in power systems, and solves the problem of unreasonable layout of battery swap stations caused by strong randomness and complex constraints of electric vehicle charging and swap stations; the present invention uses a Monte Carlo simulation method to predict the charging load of electric vehicles, so that the charging load model can reflect the uncertainty of electric vehicles connecting to battery swap stations, which is more in line with the actual situation. With the minimum capacity-load ratio of distribution network transformers and lines, the minimum loss of distribution network and the minimum voltage deviation of distribution network as the goals, a multi-objective model for optimal planning of hybrid charging and swap stations for electric vehicles is established; a particle swarm algorithm is used to solve the multi-objective model; a model comprehensively considering the random load characteristics of electric vehicles can describe the randomness of the load after the electric vehicle is connected, optimize the layout of battery swap stations according to the carrying capacity of the power grid, and more comprehensively and accurately reflect the state of the power grid.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of optimized layout of battery swap stations in an electric power system, and relates to a layout method of battery swap stations that takes into account the battery swapping needs of electric vehicles and the carrying capacity of a power grid. Background Art

[0002] With the large-scale disorderly access of charging and discharging loads such as electric vehicles and battery swap stations, the power generation and load demand on the medium and low voltage distribution side are gradually unbalanced, and the distribution network is changing to load-intensive and power-intensive. The bidirectional power flow characteristics of distribution network transformers and lines are obvious. Some areas have already seen problems such as serious distribution network overload and high operating risks. By conducting an assessment of the carrying capacity of charging and discharging loads connected to the distribution network, quantifying the evaluation indicators of the distribution network's ability to accept charging and discharging loads, and dividing the distribution network's carrying capacity level, we can provide a basis for the planning, construction, and operation of charging and discharging loads such as electric vehicles / battery swap stations and power grids, scientifically guide the reasonable access of charging and discharging loads, and promote the healthy and orderly development of the electric vehicle industry.

[0003] Many scholars at home and abroad have studied the large number of renewable distributed generation DG (Distributed Generation) and electric vehicles EV (Electric Vehicle) connected to the power grid. By establishing a time distribution model of disorderly distributed electric vehicle charging load, the impact of the distribution network after the electric vehicle charging load is connected is explored. The early warning, decision-making and visualization methods in the distribution network carrying capacity assessment play an important role in achieving the safe and stable operation of the smart grid. However, the current application of carrying capacity assessment in the distribution system is mainly aimed at the early warning and monitoring of the power supply and consumption of the distribution network. How to use the carrying capacity assessment of the regional power grid for large-scale electric vehicles / battery swap stations and other uncertain loads has not been deeply developed. Summary of the invention

[0004] The technical solution of the present invention is used to solve the problem of unreasonable layout of electric vehicle charging and swapping stations due to strong randomness and complex constraints.

[0005] The present invention solves the above technical problems through the following technical solutions:

[0006] A method for arranging battery swap stations taking into account the battery swapping demand of electric vehicles and the carrying capacity of a power grid comprises the following steps:

[0007] S1. Predict the charging load of electric vehicles using Monte Carlo simulation method based on the charging data of electric vehicles;

[0008] S2. Establish a multi-objective model for the optimal planning of electric vehicle swap stations, with the goals of maximizing the degree of user capture by electric vehicle swap stations, minimizing the capacity-load ratio of distribution network transformers and lines, minimizing distribution network losses, and minimizing distribution network voltage deviation;

[0009] S3. Use particle swarm algorithm to solve the multi-objective model and obtain the optimal solution set.

[0010] Furthermore, the method for predicting the charging load of an electric vehicle using the Monte Carlo simulation method described in step S1 is specifically as follows:

[0011] S11. Based on the distribution characteristics of the initial state of charge of the electric vehicle battery and the distribution characteristics of the charging start time, a charging probability model of the electric vehicle at a certain time t within 1 day is established as follows:

[0012] (1)

[0013] in, represents the expected value of SOC probability, represents the probability density function of SOC, represents the distribution characteristics of the charging start time, R=L / I, L represents the corresponding charging duration under a specific charging method, I=24 / x, and one day is divided into x time periods, then R represents the number of previous time points that have an impact on the charging probability at any time t, and q=0,1,2,…R is used to accumulate and calculate the impact of the time before time t on the charging probability;

[0014] S12. For the probability model, the Monte Carlo probability sampling algorithm is used to solve it. The specific process is as follows:

[0015] S121, setting the number of sampling times to m;

[0016] S122, perform sampling simulation on the initial state of charge of the battery to obtain its probability expectation I soc ;

[0017] S123, performing sampling simulation on the charging start time, obtaining the corresponding charging duration L under the specific charging mode and calculating R;

[0018] S124, substituting the sampling results of the above two steps into formula (1) to calculate the probability p(t) of battery charging at each time in a day;

[0019] S125. Calculate the expected charging power P of a single vehicle based on the charging data of the specific application scenario;

[0020] S13. Combined with the historical data of battery swap stations, all electric vehicles that have been charged at battery swap stations in the region are marked as electric vehicles in the region, leaving sufficient margin for distribution network voltage control and predicting the number of electric vehicles that need battery swapping. , charging load of battery swap station ;

[0021] S14. Determine the maximum battery capacity n based on the scale of a single battery swap station, and calculate the number of battery swap stations required in the entire area N = [S total / n]+1, the total number of electric vehicles with battery swapping needs in the entire area is divided by the maximum battery capacity n of a single battery swapping station, and rounded up to get the number N of battery swapping stations that meet the battery swapping needs in the area, ensuring that there is sufficient capacity for electric vehicles with charging needs in the area at the same time.

[0022] Furthermore, the method for establishing a multi-objective model for optimal planning of electric vehicle swap stations described in step S2 with the goals of maximizing the degree of user capture by electric vehicle swap stations, minimizing the capacity-load ratio of distribution network transformers and lines, minimizing distribution network losses, and minimizing distribution network voltage deviation is as follows:

[0023] S21. Calculate the driving distance of users who need battery swapping. Divide the electric vehicles in the area to be replaced into i areas. The distance cost from the user to the access node battery swapping station is given by the following formula:

[0024] (2)

[0025] Among them, i is the network node; j is the network node connected to the battery swap station; d ij is the spatial distance from the battery swapping demand point i to the battery swapping station j; S total,i The number of electric vehicles for battery replacement is predicted on the network node i;

[0026] S22. Define the minimum objective function of the capacity-load ratio of the grid transformer and line as follows:

[0027] (3)

[0028] Among them, R i Indicates whether a battery swap station is built at network node i, R i 1 means building a charging and swapping station at node i, R i =0, indicating that no charging and swapping station is built at network node i; P si , P Ri are the transformer and line capacity ratios of network node i respectively;

[0029] S23, calculate the capacity load ratio, the specific steps are as follows:

[0030] S231, determining a voltage level or a power supply area, and collecting data on factors affecting the capacity-load ratio of a certain voltage level or power supply area;

[0031] S232. Based on the sum of the public distribution transformer capacities and the hierarchical loads within the power supply area of ​​the medium voltage distribution network in the data, a mathematical model of the medium voltage capacity-to-load ratio CUR is established as follows:

[0032] (4)

[0033] in, is the actual load power of the distribution network; It is the capacity of the distribution network, that is, the maximum rated power of the transformer or other equipment.

[0034] S233. Calculate the maximum load of the 220kV and below network and the maximum load of the 10kV distribution transformer within the jurisdiction based on the collected data; the actual load power of the distribution network Calculation based on load curve or actual measurement data; capacity of distribution network Determined by the equipment in the distribution network, assuming that the distribution network consists of N transformers and M lines, the total capacity is expressed as:

[0035] (5)

[0036] in, is the capacity of the i-th transformer; is the capacity of the jth line;

[0037] S234, based on the calculated maximum load, using the medium pressure capacity-to-load ratio mathematical model to calculate the medium pressure capacity-to-load ratio, calculating the medium pressure capacity-to-load ratio difference and selecting an error range to obtain a final medium pressure capacity-to-load ratio CUR;

[0038] S24, the objective function of minimizing the loss of the distribution network is as follows:

[0039] (6)

[0040] Among them, R i Indicates whether a battery swap station is built at network node i, R i The value is 1, which means that a battery swap station is built at node i. i The value of 0 means that the network node i does not have a battery swap station; li , Q li are the active and reactive capacities of the battery swap station at network node i respectively;

[0041] S25, distribution network voltage deviation

[0042] The power flow can be characterized by node voltage and branch power. The impact of the charging load of the electric vehicle charging station on the power flow of the distribution network is approximately evaluated by the node voltage offset of the distribution network described by formula (7):

[0043] (7)

[0044] Among them, u max and u min They represent the maximum and minimum voltage amplitudes of all nodes in the power distribution network connected to the battery swap station; u bIndicates the voltage reference value of the distribution network;

[0045] S26. Constraints

[0046] S.61, battery quantity constraint of the battery swap station. Since the actual number of batteries that can be provided by the battery swap station is limited, the number of batteries n in the optimization process is restricted as follows:

[0047] (8)

[0048] Among them, K is the maximum battery capacity designed for the battery swap station;

[0049] S262, the power flow balance equation of each node is:

[0050] (9)

[0051] in, is the injected active power of node i, is the injected reactive power of node i; is the voltage at node i; is the conductance between node i and node j, reflecting the active power transfer capability of the conductive path; is the phase angle difference between node i and node j, is the imaginary part of the admittance between node i and node j, Represents a special node that can control active or reactive power;

[0052] S263, node voltage constraint is:

[0053] (10)

[0054] in, , are the upper and lower limits of the voltage amplitude of node i respectively;

[0055] S264, the power constraint of the battery swap station is:

[0056] (11)

[0057] in, is the reactive power of reactive power compensation device i, , They are respectively the upper and lower limits of the reactive power capacity of the reactive compensation device.

[0058] Furthermore, the method of using the particle swarm algorithm to solve the multi-objective model in step S3 to obtain the optimal solution set is as follows:

[0059] S31, initialize a particle swarm of size N, set various boundary conditions of the particles, randomly assign initial positions and speeds to each particle under the control variable constraint conditions, substitute the initial positions and speeds into the objective function to calculate the corresponding fitness values, and update the flight speed and position through the position update formula according to the optimal position searched;

[0060] S32, guide the extrapolation direction through the fitness value difference of particles, combine with random numbers, and use formula , in the particle Generates a virtual location nearby , and get the next particle Virtual Location , through sorting:

[0061] (12)

[0062] in, is the virtual position of the i-th particle at time t+1; For random numbers, randomness is introduced to increase diversity and prevent the particle swarm from falling into the local optimal uniform distribution between [0,1]; is the current position of the i-th particle in the search space at time t; is the current position of the i-th particle in the search space at time t+1, Used to control the search speed;

[0063] S33, during the iterative calculation process, determine whether the particle is in premature state. If so, add an adaptive ε amplitude fine-tuning operator after formula (12), as follows:

[0064] (13)

[0065] S34. Calculate using the objective function , The fitness value corresponding to the particle, re-evaluate the fitness of each particle, retain the optimal solution, and update the individual optimal P i and the global optimal P g .

[0066] An electronic device includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the above-mentioned battery replacement station layout method that takes into account the battery replacement needs of electric vehicles and the carrying capacity of the power grid, and the processor is configured to execute the program stored in the memory.

[0067] A storage medium stores a computer program, which, when executed by a processor, executes the steps of the above-mentioned method for laying out battery swap stations taking into account the battery swapping needs of electric vehicles and the carrying capacity of a power grid.

[0068] The advantages of the present invention are:

[0069] The present invention firstly predicts the charging load of electric vehicles by using the Monte Carlo simulation method according to the rental rules of electric vehicles and charging data such as power and time, so that the charging load model can reflect the uncertainty of electric vehicles connecting to battery swap stations, which is more in line with the actual situation; secondly, a multi-objective model for optimal planning of hybrid charging and swap stations for electric vehicles is established with the goals of minimizing the capacity-load ratio of distribution network transformers and lines, minimizing distribution network losses and minimizing distribution network voltage deviation; finally, the multi-objective model is solved by using a particle swarm algorithm; the present invention establishes a model that comprehensively considers the random load characteristics of electric vehicles, so that the randomness of the load after the electric vehicle is connected can be described, the layout of the battery swap station is optimized according to the carrying capacity of the power grid, and the state of the power grid can be more comprehensively and accurately reflected. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 A flow chart of a method for laying out battery swap stations taking into account the battery swapping needs of electric vehicles and the carrying capacity of a power grid according to the present invention;

[0071] Figure 2 This is a schematic diagram of the IEEE 69 node distribution network;

[0072] Figure 3 A comparison diagram of the optimized layout of battery swap stations in an IEEE 69-node power distribution network using the battery swap station layout method of the present invention that takes into account the battery swap demand of electric vehicles and the carrying capacity of the power grid;

[0073] Figure 4 This is a load comparison diagram of the optimized layout of battery swap stations in the IEEE 69-node power distribution network using the battery swap station layout method of the present invention that takes into account the battery swapping needs of electric vehicles and the carrying capacity of the power grid. DETAILED DESCRIPTION

[0074] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in combination with the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0075] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and specific embodiments:

[0076] Embodiment 1

[0077] like Figure 1 As shown, a method for optimizing the layout of battery swap stations in this embodiment considering the battery swap demand of electric vehicles and the carrying capacity of the power grid specifically includes the following steps:

[0078] Step 1: Forecast the charging load of electric vehicles

[0079] The load forecasting model that considers the impact of electric vehicle battery replacement demand calculated by the Monte Carlo probability sampling algorithm can more accurately and objectively reflect the uncertainty of information, improve the accuracy of random load prediction after electric vehicles are connected, and effectively reflect various uncertain information such as randomness, grayness, and fuzziness. The present invention uses a probability model to simulate charging behavior. First, the historical data of the battery replacement station and the actual application data of the car of the car company operator are obtained (usually the data is recorded once an hour and stored separately according to the date), which at least includes the following data: charging times, charging time, charging station location, charging power, charging status, mileage, and fuel price. The data of four time periods, namely, the valley period, the flat period, the peak period, and the weekend period, are selected as representative data. The charging load characteristics of electric vehicles in the four time periods can cover most charging scenarios. In addition, a certain random time period data fragment is added to the data to simulate the random factors that affect the charging load of electric vehicles, and finally an electric vehicle charging probability model is established, and the Monte Carlo probability sampling algorithm is used to solve it.

[0080] Step 1.1: Based on the distribution characteristics of the initial state of charge of electric vehicle batteries and the distribution characteristics of the charging start time, the charging probability model of electric vehicles at a certain time t within 1 day is established as follows:

[0081] (1)

[0082] in, represents the expected probability value of SOC (between 0 and 1), The probability density function of SOC is expressed by collecting a large amount of SOC data of electric vehicles and analyzing their distribution characteristics. SOC can usually be expressed by a probability density function In other words, we use a probability density function Represents the distribution characteristics of the charging start time; R=L / I, L represents the corresponding charging duration under a specific charging method, I=24 / x, and 1 day is divided into x time periods, then R represents the number of previous time points that affect the charging probability at any time t, and q=0,1,2,…R is used to accumulate and calculate the impact of the time before time t on the charging probability, and then select an appropriate probability model based on the data, such as normal distribution, Beta distribution, etc.

[0083] Step 1.2: For the probability model, use the Monte Carlo probability sampling algorithm to solve

[0084] Step 1.2.1, set the sampling times m according to the calculation accuracy;

[0085] Step 1.2.2: According to the distribution characteristics of the average daily mileage of the vehicle and the relevant initial settings, a sampling simulation is performed on the initial state of charge of the battery to obtain its probability expectation I soc ;

[0086] Step 1.2.3: Based on the vehicle travel statistics and the actual application scenario, a sampling simulation is performed on the charging start time to obtain its expected probability L and calculate R;

[0087] Step 1.2.4: Substitute the sampling results of the above two steps into equation (1) to calculate the expected probability p(t) of EV charging at each time in a day.

[0088] Step 1.2.5: Calculate the expected charging power P of a single vehicle based on the charging data of the specific application scenario.

[0089] Step 1.3: Combined with the historical data of battery swap stations, all electric vehicles that have been charged at battery swap stations in the area are marked as electric vehicles in the area, so as to leave sufficient margin for distribution network voltage control and predict the number of electric vehicles that need battery swapping. , charging load of battery swap station The charging power is determined by the charging method in the actual application scenario corresponding to a specific type of vehicle and is related to the battery charging characteristics. In general, the dynamic response of the battery charging characteristics is ignored and the charging power is approximately considered to be a constant value.

[0090] Step 1.4: Determine the maximum battery capacity n based on the scale of a single battery swap station, and further calculate the number of battery swap stations required in the entire area N = [S total / n]+1, the total number of electric vehicles with battery swapping needs in the entire area is divided by the maximum battery capacity n of a single battery swapping station, and rounded up to get the number N of battery swapping stations that meet the battery swapping needs in the area, ensuring that there is sufficient capacity for electric vehicles with charging needs in the area at the same time.

[0091] Step 2: Establish a multi-objective model for optimal planning of electric vehicle battery swap stations

[0092] Step 2.1, calculate the driving distance of users who need battery swapping, divide the electric vehicles in the area to be replaced into i areas, and in order to ensure that the battery swapping station is as convenient as possible for users to swap batteries, the distance cost from users to the access node battery swapping station is given by the following formula:

[0093] (2)

[0094] Among them, i is the network node; j is the network node connected to the battery swap station; d ij is the spatial distance from the battery swapping demand point i (network node) to the battery swapping station j; S total,i is the predicted number of electric vehicles to be replaced by batteries at network node i.

[0095] Step 2.2: Define the minimum objective function of the capacity-load ratio of distribution network transformers and lines as follows:

[0096] (3)

[0097] Among them, R i Indicates whether a battery swap station is built at network node i, R i 1 means building a charging and swapping station at node i, R i =0, indicating that no charging and swapping station is built at network node i; P si , P Ri are the transformer and line capacity ratios of network node i respectively.

[0098] Step 2.3: Calculate the load capacity ratio

[0099] Step 2.3.1, determine the voltage level or power supply area, and collect data on factors affecting the capacity-load ratio of a certain voltage level or power supply area;

[0100] Step 2.3.2: Based on the sum of the public distribution transformer capacity and the hierarchical load within the power supply area of ​​the medium voltage distribution network in the data, a mathematical model of the medium voltage capacity utilization ratio (CUR) is established as follows:

[0101] (4)

[0102] in, is the actual load power of the distribution network; It is the capacity of the distribution network, that is, the maximum rated power of the transformer or other equipment.

[0103] Step 2.3.3: Calculate the maximum load of the 220kV and below network and the maximum load of the 10kV distribution transformer in the area based on the collected data; the actual load power of the distribution network Calculation based on load curve or actual measurement data; capacity of distribution network It is determined by the equipment in the distribution network (such as transformers, lines, etc.). Assuming that the distribution network consists of N transformers and M lines, the total capacity is expressed as:

[0104] (5)

[0105] in, is the capacity of the i-th transformer; is the capacity of the jth line.

[0106] Step 2.3.4, based on the calculated maximum load, use the mathematical model to calculate the medium pressure capacity-load ratio, calculate the medium pressure capacity-load ratio difference and select the error range to obtain the final capacity-load ratio CUR;

[0107] Step 2.4, consider the grid network loss as the objective function as follows:

[0108] (6)

[0109] Among them, R i Indicates whether a battery swap station is built at network node i, R i The value is 1, which means that a battery swap station is built at node i. i The value of 0 means that the network node i does not have a battery swap station; li , Q li are the active and reactive capacities of the battery swap station at network node i respectively.

[0110] Step 2.5: Consider voltage deviation

[0111] The power flow can be characterized by node voltage and branch power. The impact of the charging load of the EV charging station on the power flow of the distribution network can be approximately evaluated by the node voltage deviation of the distribution network described by formula (7):

[0112] (7)

[0113] Among them, u max and u min They represent the maximum and minimum voltage amplitudes of all nodes in the power distribution network connected to the battery swap station; u b Indicates the voltage reference value of the distribution network.

[0114] Step 2.6: Constraints

[0115] Step 2.6.1: Constraint on the number of batteries in the battery swap station. Since the number of batteries that can actually be provided by the battery swap station is limited, the number of batteries n in the optimization process is restricted:

[0116] (8)

[0117] Among them, K is the maximum battery capacity designed for the battery swap station;

[0118] Step 2.6.2, Equality constraint, the equality constraint is the power flow balance equation of each node. Since there are two ways to express node voltage - polar coordinates and rectangular coordinates, polar coordinates are used here, that is:

[0119] (9)

[0120] in, is the injected active power of node i, is the injected reactive power of node i; is the voltage at node i; is the conductance between node i and node j, reflecting the active power transfer capability of the conductive path; is the phase angle difference between node i and node j, is the imaginary part of the admittance (susceptance) between node i and node j, Represents a special node that can control active or reactive power.

[0121] Step 2.6.3, inequality constraints, node voltage constraints:

[0122] (10)

[0123] in, , are the upper and lower limits of the voltage amplitude of node i respectively.

[0124] Step 2.6.4, inequality constraints, power constraints of battery swap stations:

[0125] (11)

[0126] in, is the reactive power of reactive power compensation device i, , They are respectively the upper and lower limits of the reactive power capacity of the reactive compensation device.

[0127] It should be noted that the variables for reactive power optimization in this chapter are all continuous variables, and discrete variables (such as transformer tap ratio, number of parallel capacitors or reactors, etc.) are not considered.

[0128] Step 3: Use particle swarm algorithm to solve the multi-objective model

[0129] Step 3.1: First, initialize a particle swarm of size N, set the boundary conditions of the particles, randomly assign initial positions and velocities to each particle while satisfying the control variable constraints, substitute the objective function to calculate the corresponding fitness value, and update its own flight speed and position according to the optimal position searched through the position update formula in the basic particle swarm algorithm.

[0130] Step 3.2: Use the fitness value difference of the particles to guide the extrapolation direction, combine with random numbers, and use the formula , in the particle Generates a virtual location nearby , and get the next particle Virtual Location , through sorting:

[0131] (12)

[0132] in, is the virtual position of the i-th particle at time t+1; For random numbers, randomness is introduced to increase diversity and prevent the particle swarm from falling into the local optimal uniform distribution between [0,1]; is the current position of the i-th particle in the search space at time t; is the current position of the i-th particle in the search space at time t+1, Used to control the search speed;

[0133] Step 3.3: In the iterative calculation process, if some components of adjacent particles are similar or even the same, it is judged that the particles are premature, and an adaptive ε amplitude fine-tuning operator is added after formula (12):

[0134] (13)

[0135] Step 3.4: Calculate using the objective function , The fitness value corresponding to the particle, re-evaluate the fitness of each particle, retain the optimal solution, and update the individual optimal P i and the global optimal P g .

[0136] Test verification and analysis

[0137] Taking an IEEE 69-node power distribution network as an example, the optimal access point of the battery swap station is calculated.

[0138] The present invention takes into account the user-side needs and grid requirements. The objective function includes four sub-goals: the cost of the battery replacement distance, the capacity-load ratio of the grid transformer and the line, the grid network loss, and the voltage deviation. To verify the feasibility of the method of the present invention, the following are selected: Figure 2 Taking the IEEE 69-node power distribution network shown in the figure as an example (the nodes in the figure do not represent the spatial position), the best access point of the battery swap station is calculated, and the voltage fluctuations, user satisfaction, loss comparison, etc. of different access points are compared.

[0139] Firstly, based on the geographical location coupling of the IEEE 69-node distribution network and the corresponding road traffic network nodes, the simulation calculates the minimum number of battery swap station nodes N=7 according to the total number of electric vehicles in the region and the number of batteries that the battery swap station can accommodate.

[0140] A set of comparisons is set. The conventional battery swap station access method sets a battery swap station access node for every 10 nodes (the node indicated by the triangle in the figure). The optimized battery swap station access node is marked with a five-pointed star according to the algorithm calculation result. The battery swap station node setting is as follows: Figure 3 shown.

[0141] The corresponding parameter comparison before and after applying the optimization method is shown in Table 1.

[0142] Table 1 Comparison of results under different control strategies

[0143]

[0144] It can be seen from the data in Table 1 that after optimization, without changing the capacity-load ratio of the grid transformer and the line, the battery swap distance cost in the distribution network has dropped from 1287.6 in the non-optimized layout to 1226.4, and the battery swap distance cost has dropped by about 4.8%; thanks to the reasonable layout of battery swap stations, the network loss of the distribution network has also been reduced; the voltage deviation caused by the uneven battery swap demand in the original battery swap station layout has also been reduced after the optimized layout.

[0145] Figure 4 The figure shows the changes in the load rate of each node after the optimized layout. It can be clearly seen that the load rate of the nodes with higher original load rates has decreased. After optimization, part of the electric vehicle load in the high-load area is allocated to the low-load area, which increases the load rate of the nodes with lower original load rates and ultimately improves the overall load rate.

[0146] Embodiment 2

[0147] An electronic device includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the method for arranging battery swap stations that considers the battery swapping needs of electric vehicles and the carrying capacity of a power grid in Example 1, and the processor is configured to execute the program stored in the memory.

[0148] Embodiment 3

[0149] A storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of the method for laying out battery swap stations taking into account the battery swapping demand of electric vehicles and the carrying capacity of a power grid in Embodiment 1 are executed.

[0150] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for arranging battery swap stations taking into account the battery swapping needs of electric vehicles and the carrying capacity of the power grid, characterized in that: The following steps are involved: S1. Based on the charging data of electric vehicles, the charging load of electric vehicles is predicted using the Monte Carlo simulation method. The specific method is as follows: S11. Based on the distribution characteristics of the initial state of charge of the electric vehicle battery and the distribution characteristics of the charging start time, a charging probability model of the electric vehicle at a certain time t within 1 day is established as follows: (1) in, represents the expected value of SOC probability, represents the probability density function of SOC, represents the distribution characteristics of the charging start time, R=L / I, L represents the charging duration, I=24 / x, and one day is divided into x time periods, then R represents the number of previous time points that have an impact on the charging probability at any time t, and q=0,1,2,…R is used to accumulate and calculate the impact of the time before time t on the charging probability; S12. For the probability model, the Monte Carlo probability sampling algorithm is used to solve it. The specific process is as follows: S121, setting the number of sampling times to m; S122, perform sampling simulation on the initial state of charge of the battery to obtain its probability expectation I soc ; S123, performing sampling simulation on the charging start time, obtaining the charging duration L and calculating R; S124, substituting the sampling results of the above two steps into formula (1) to calculate the probability p(t) of battery charging at each time in a day; S125. Calculate the expected charging power P of a single vehicle based on the charging data of the specific application scenario; S13. Combined with the historical data of battery swap stations, all electric vehicles that have been charged at battery swap stations in the region are marked as electric vehicles in the region, leaving sufficient margin for distribution network voltage control and predicting the number of electric vehicles that need battery swapping. , charging load of battery swap station ; S14. Determine the maximum battery capacity n based on the scale of a single battery swap station, and calculate the number of battery swap stations required in the entire area N = [S total / n]+1, the total number of electric vehicles with battery swapping needs in the entire area is divided by the maximum battery capacity of a single battery swapping station n, and rounded up to get the number of battery swapping stations N that meet the battery swapping needs in the area, ensuring that there is sufficient capacity for electric vehicles with charging needs in the area at the same time; S2. A multi-objective model for the optimal planning of electric vehicle swap stations is established with the goals of maximizing the degree of user capture by electric vehicle swap stations, minimizing the capacity-load ratio of distribution network transformers and lines, minimizing distribution network losses, and minimizing distribution network voltage deviation. The specific method is as follows: S21. Calculate the driving distance of users who need battery swapping. Divide the electric vehicles in the area to be replaced into i areas. The distance cost from the user to the access node battery swapping station is given by the following formula: (2) Among them, i is the network node; j is the network node connected to the battery swap station; d ij is the spatial distance from the battery swapping demand point i to the battery swapping station j; S total,i The number of electric vehicles for battery replacement is predicted on the network node i; S22. Define the minimum objective function of the capacity-load ratio of the grid transformer and line as follows: (3) Among them, R i Indicates whether a battery swap station is built at network node i, R i 1 means building a charging and swapping station at node i, R i =0, indicating that no charging and swapping station is built at network node i; P si , P Ri are the transformer and line capacity ratios of network node i respectively; S23, calculate the capacity load ratio, the specific steps are as follows: S231, determining a voltage level or a power supply area, and collecting data on factors affecting the capacity-load ratio of a certain voltage level or power supply area; S232. Based on the sum of the public distribution transformer capacities and the hierarchical loads within the power supply area of ​​the medium voltage distribution network in the data, a mathematical model of the medium voltage capacity-to-load ratio CUR is established as follows: (4) in, is the actual load power of the distribution network; is the capacity of the distribution network, i.e. the maximum rated power of the transformer or other equipment; S233. Calculate the maximum load of the 220kV and below network and the maximum load of the 10kV distribution transformer within the jurisdiction based on the collected data; the actual load power of the distribution network Calculation based on load curve or actual measurement data; capacity of distribution network Determined by the equipment in the distribution network, assuming that the distribution network consists of N transformers and M lines, the total capacity is expressed as: (5) in, is the capacity of the i-th transformer; is the capacity of the jth line; S234, based on the calculated maximum load, using the medium pressure capacity-to-load ratio mathematical model to calculate the medium pressure capacity-to-load ratio, calculating the medium pressure capacity-to-load ratio difference and selecting an error range to obtain a final medium pressure capacity-to-load ratio CUR; S24, the objective function of minimizing the loss of the distribution network is as follows: (6) Among them, R i Indicates whether a battery swap station is built at network node i, R i The value is 1, which means that a battery swap station is built at node i. i The value of 0 means that the network node i does not have a battery swap station; li , Q li are the active and reactive capacities of the battery swap station at network node i respectively; S25, distribution network voltage deviation The power flow can be characterized by node voltage and branch power. The impact of the charging load of the electric vehicle charging station on the power flow of the distribution network is approximately evaluated by the node voltage offset of the distribution network described by formula (7): (7) Among them, u max and u min They represent the maximum and minimum voltage amplitudes of all nodes in the power distribution network connected to the battery swap station; u b Indicates the voltage reference value of the distribution network; S26. Constraints S261, battery quantity constraint of the battery swap station. Since the actual number of batteries that can be provided by the battery swap station is limited, the number of batteries n in the optimization process is restricted as follows: (8) Among them, K is the maximum battery capacity designed for the battery swap station; S262, the power flow balance equation of each node is: (9) in, is the injected active power of node i, is the injected reactive power of node i; is the voltage at node i; is the conductance between node i and node j, reflecting the active power transfer capability of the conductive path; is the phase angle difference between node i and node j, is the imaginary part of the admittance between node i and node j, Represents a special node that can control active or reactive power; S263, node voltage constraint is: (10) in, , are the upper and lower limits of the voltage amplitude of node i respectively; S264, the power constraint of the battery swap station is: (11) in, is the reactive power of reactive power compensation device i, , They are the upper and lower limits of the reactive power capacity of the reactive compensation device respectively; S3. Use particle swarm algorithm to solve the multi-objective model and obtain the optimal solution set.

2. The method for arranging battery swap stations considering the battery swapping demand of electric vehicles and the carrying capacity of the power grid according to claim 1 is characterized in that: The method of using the particle swarm algorithm to solve the multi-objective model and obtain the optimal solution set described in step S3 is as follows: S31, initialize a particle swarm of size N, set various boundary conditions of the particles, randomly assign initial positions and speeds to each particle under the control variable constraint conditions, substitute the initial positions and speeds into the objective function to calculate the corresponding fitness values, and update the flight speed and position through the position update formula according to the optimal position searched; S32, guide the extrapolation direction through the fitness value difference of particles, combine with random numbers, and use formula , in the particle Generates a virtual location nearby , and get the next particle Virtual Location , through sorting: (12) in, is the virtual position of the i-th particle at time t+1; For random numbers, randomness is introduced to increase diversity and prevent the particle swarm from falling into the local optimal uniform distribution between [0,1]; is the current position of the i-th particle in the search space at time t; is the current position of the i-th particle in the search space at time t+1, Used to control the search speed; S33, during the iterative calculation process, determine whether the particle is in premature state. If so, add an adaptive ε amplitude fine-tuning operator after formula (12), as follows: (13) S34. Calculate using objective function , The fitness value corresponding to the particle, re-evaluate the fitness of each particle, retain the optimal solution, and update the individual optimal P i and the global optimal P g .

3. An electronic device, comprising a memory and a processor, characterized in that: The memory is used to store a program that supports the processor to execute the battery swap station layout method that takes into account the battery swapping needs of electric vehicles and the carrying capacity of the power grid as described in any one of claims 1 to 2, and the processor is configured to execute the program stored in the memory.

4. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the steps of the method for laying out battery swap stations taking into account the battery swapping needs of electric vehicles and the carrying capacity of the power grid as described in any one of claims 1 to 2 are executed.