Method for calculating carrying capacity of electric vehicle in power distribution network

Abstract

The invention discloses a power distribution network electric vehicle bearing capacity calculation method, and belongs to the technical field of power distribution network bearing capacity calculation. The method for calculating the carrying capacity of the electric vehicle in the power distribution network comprises the following steps: constructing a trip chain model of the electric vehicle, and describing transfer and stop of the electric vehicle between different places through a state transfer matrix; continuously simulating by adopting a Monte Carlo simulation method to obtain the charging requirements of the electric vehicles in each region in each time period of the whole day; constructing a power distribution network electric vehicle bearing capacity optimization model with the maximum access capacity; a Wasserstein distance is adopted to construct a fuzzy set of photovoltaic output, a distributed robust optimization model is established, and the optimization model is converted for solving. The method for calculating the carrying capacity of the electric vehicle in the power distribution network can solve the problem of poor calculation accuracy of the carrying capacity of the electric vehicle in the power distribution network in the existing calculation method, and has the advantage of ensuring the safe and stable operation of the power distribution network after the charging load is accessed.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network carrying capacity calculation, and in particular to a method for calculating the carrying capacity of electric vehicles in a distribution network. Background Art

[0002] Currently, a large number of new energy electric vehicles are connected to distribution networks. However, the charging load of electric vehicles is characterized by randomness, volatility, intermittence, and mobility. This large-scale, disorderly connection may overlap with the peak hours of conventional loads, resulting in a "peak upon peak" situation, affecting the safe and stable operation of the distribution network. Therefore, under the premise of meeting node voltage, line capacity, and other safe operation constraints, evaluating the maximum EV connection capacity of the distribution network—that is, the distribution network's EV carrying capacity—is crucial for the safe and stable operation of the distribution network.

[0003] Methods for calculating the electric vehicle carrying capacity of distribution networks primarily include simulation-based methods and optimization model-based methods. The simulation-based calculation method typically involves building a distribution network model using power flow simulation software and gradually increasing the penetration rate of electric vehicles in the distribution network until the relevant constraint indicators exceed the limit. This method is relatively simple in principle, but its universality is limited, and the use of electric vehicle penetration as an indicator of carrying capacity is relatively crude. Optimization model-based methods currently rarely consider the volatility of photovoltaic power output connected to the distribution network when calculating electric vehicle carrying capacity. This is the impact of the uncertainty of photovoltaic output on the electric vehicle carrying capacity of the distribution network, resulting in poor accuracy in the calculated distribution network load carrying capacity. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for calculating the carrying capacity of electric vehicles in a distribution network, which solves the problem of poor accuracy of the existing calculation methods for the carrying capacity of electric vehicles in a distribution network and has the advantage of ensuring the safe and stable operation of the distribution network after connecting to the charging load.

[0005] To achieve the above object, the present invention provides a method for calculating the carrying capacity of electric vehicles in a distribution network, comprising the following steps: S1. Analyze the statistics of electric vehicle travel to obtain the travel patterns of electric vehicle users, build an electric vehicle travel chain model, and describe the transfer and stay of electric vehicles between different locations through the state transition matrix; S2. Monte Carlo simulation is used to continuously simulate the charging demand of electric vehicles in various areas at various times of the day; S3. Construct an optimization model for electric vehicle carrying capacity of the distribution network with the largest access capacity; S4. Obtain the historical data sample set of photovoltaic output, construct the fuzzy set of photovoltaic output using the Wasserstein distance, establish a distributionally robust optimization model, and transform the optimization model containing the uncertainty of photovoltaic output into a deterministic electric vehicle carrying capacity model for solution.

[0006] Preferably, in the said S1, the travel chain describes the whole process that the user starts from the starting point based on the travel purpose, passes through several passing locations in sequence according to the time order, and finally arrives at the destination to complete the travel. The expression of the state transition matrix is: ; Among them, represents T the probability of going from area M to area N at time M is H , W , P or O , N is H , W , P or O , H represents the residential area, W represents the work and business area, P represents the public entertainment area, O represents other areas.

[0007] Preferably, in the said S1, after the electric vehicle completes the journey, the time it stays at different destination nodes is fitted by the probability density function. The probability density function of the parking duration of the electric vehicle in the residential area is: ; Among them, is the shape parameter, is the scale parameter, is the parking duration of the electric vehicle; The probability density function of the parking duration of the electric vehicle in W , P , O area is: ; Among them, is the location parameter, is the scale parameter, is the shape parameter; when , ; when , .

[0008] Preferably, in S2, the specific process of continuously simulating the charging demands of electric vehicles in each region at each time period of the whole day by using the Monte Carlo simulation method is as follows: S21. Model the road network to generate a road distance matrix D, and input the road distance matrix and the road impedance matrix calculated according to the road impedance model; S22. Construct an electric vehicle charging load model, initialize the basic vehicle information, extract the complete travel chain of each electric vehicle, and plan the travel path with the shortest travel time for the vehicle based on the road impedance matrix according to the travel demands of the vehicle at different times; S23. Set different numbers of electric vehicles for Monte Carlo sampling simulation, repeat the continuous simulation, analyze and obtain the charging demands corresponding to each road network node at each moment, and obtain the spatio-temporal distribution of the electric vehicle charging demands.

[0009] Preferably, in S21, abstract the road network topology structure into a weighted directed graph, and the topological mathematical expression model of the road network is: ; Among them, represents the traffic road network, represents the graph the set of all nodes, represents a specific node, is the number of nodes; represents the graph the set of all road sections, represents the road section connecting node and node ; represents the set of divided time periods, is the number of divided time periods, t represents a specific time period; W is the set of road section weights, represents the weight of road section t in the time period ; The road section weight W represents the road travel cost, t At time The impedance model of the road section is: ; Among them, the saturation degree , is the traffic flow of the road section, C is the traffic capacity; t 0 is the zero-flow travel time, are all impedance influence factors.

[0010] Preferably, in said S22, a charging load model of an electric vehicle is constructed, including battery capacity, battery state of charge, charging load and charging time; The battery capacity of different types of electric vehicles obeys a uniform distribution, which can be expressed as: ; in, Indicates the battery capacity of electric vehicles, are the upper and lower limits of the electric vehicle battery capacity distribution range, respectively, in kW·h; Based on the travel chain model, the battery state of charge (SOC) of an electric vehicle at the end of each trip is: ; in, Indicates the Electric cars in the SOC at the end of the trip, For the SOC at the end of the trip, is the vehicle battery capacity; For electric vehicles Mileage of the trip; The power consumption of electric vehicles per kilometer, in kW / km; When an electric vehicle generates a charging demand, the calculation expression of the charging load and charging time is: ; in, Indicates the Electric vehicles in t The charging capacity of the target charge at all times; is a 0-1 variable, indicating the The charging demand status of electric vehicles; Target charging SOC for electric vehicles; For the Electric vehicles from t The duration of charging from the moment it starts; Indicates charging power.

[0011] Preferably, in S3, constructing an optimization model for the electric vehicle carrying capacity of a distribution network with the largest access capacity includes the following steps: S31, establishing an objective function; The specific expression of the objective function is: ; Among them, the collection Indicates the functional area to which EVCS belongs, . For functional areas The electric vehicle charging load that the nth node can access, is the charging demand of the functional area predicted by simulation, is the number of nodes with charging stations in the functional area , is the parameter that controls the steepness of the satisfaction function, is the parameter that represents the influence degree of the penalty term on the charging satisfaction. The operator max{X,0} means taking the larger value of the two, that is, if X≥0, the value is X, otherwise it is 0.

[0012] S32. Construct constraint conditions, including node voltage constraints, branch current constraints, branch transmission power constraints, power flow constraints, and charging demand constraints.

[0013] Preferably, in the above S32, The expression of the node voltage constraint is: ; wherein, is the voltage of node , and are the upper and lower limits of the node voltage respectively, is the set of nodes.

[0014] The expression of the branch current constraint is: ; wherein, is the current of branch , is the upper limit of the branch current, is the set of branches.

[0015] The branch transmission power constraint is: ; wherein, and are the active power and reactive power flowing through branch respectively; is the maximum apparent power flowing through branch . Converting it into a linearized form, the expression is: ; The power flow constraint adopts the Distflow model.

[0016] Preferably, in the above S4, the construction process of the fuzzy set is specifically as follows: Construct an empirical probability distribution based on the historical data of the uncertainty variable, and its specific expression is: ; Among them, is the historical empirical distribution, represents the Dirac distribution centered on the historical data of the uncertainty variable ; The Wasserstein distance is used to measure the distance between two distributions, and the expression of the Wasserstein distance is: ; Among them, is the true distribution, and The support sets of are both , represents and The joint probability distribution of represents the norm; Given the radius , the data-driven Wasserstein fuzzy set is constructed as: ; Among them, represents The support set of constructs a sphere with a radius of to represent the distance between the true distribution and the historical empirical distribution .

[0017] Preferably, in the above S4, the construction and transformation process of the distributionally robust optimization model is specifically as follows: The uncertainty of the photovoltaic output is expressed as: ; Among them, is the actual photovoltaic output at node , is the predicted value of the photovoltaic output at node , is the predicted deviation value of the photovoltaic output at node ; Considering that the expectation of the function with respect to the uncertainty variable is minimized under the worst-case distribution, the optimization problem is rewritten in the following form: ;

[0018] Among them, is the decision variable, represents The feasible region, is the support set of the random variable .

[0019] According to the strong duality theorem, this problem can be transformed into the following solvable form: ;

[0020] where, The upper and lower bounds of are and is the introduced dual variable.

[0021] The advantages and positive effects of the method for calculating the carrying capacity of electric vehicles in the distribution network according to the present invention are as follows: First, based on the coupling model simulation of the road network and the distribution network, the charging demand of electric vehicles is obtained; by obtaining the travel patterns of electric vehicle users, a spatio-temporal distribution model of the charging demand is generated using Monte Carlo simulation; subsequently, an optimization model with the maximization of charging satisfaction as the goal is constructed, and considering the uncertainty of distributed photovoltaic output, a distributionally robust optimization model is constructed based on the Wasserstein distance. By solving the dual transformation of the model, the maximum accessible charging load capacity of each node in the distribution network is determined. The present invention can accurately evaluate the node load carrying capacity, effectively cope with the uncertainty of photovoltaic output, and improve the safety and reliability of the operation of the distribution network.

[0022] The technical solution of the present invention will be further described in detail below with reference to the drawings and embodiments. Description of the Drawings

[0023] Figure 1 is the flowchart of the embodiment of the calculation method of the present invention; Figure 2 is the flowchart of generating the spatio-temporal distribution of the charging load demand of the embodiment of the calculation method of the present invention; Figure 3 is the 13-node Nguyen-Dupuis road network; Figure 4 is the improved IEEE-33 node distribution network; Figure 5 is the charging load distribution curve of each functional area obtained by Monte Carlo simulation; Figure 6Graph showing the analysis results of the charging demand and electric vehicle carrying capacity of an electric vehicle charging station (EVCS); (a) Graph showing the analysis results of the charging demand and electric vehicle carrying capacity of Electric Vehicle Charging Stations 1 and 2 in Area P; (b) Graph showing the analysis results of the charging demand and electric vehicle carrying capacity of Electric Vehicle Charging Stations 3-5 in Area R; (c) Graph showing the analysis results of the charging demand and electric vehicle carrying capacity of Electric Vehicle Charging Stations 6 and 7 in Area W; (d) Graph showing the analysis results of the charging demand and electric vehicle carrying capacity of Electric Vehicle Charging Station 8 in Area O. Detailed implementation manners

[0024] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by terms such as "upper", "lower", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the product of this invention is usually placed during use. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be understood as a limitation to the present invention. In the description of the present invention, it should also be noted that unless otherwise clearly specified and defined, the terms "set", "installed", "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0025] In this application, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this application belongs. If there is any inconsistency, it shall be based on the meaning described in this specification or the meaning obtained according to the content recorded in this specification. In addition, the terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0026] The following will describe in detail the implementation manners of the present invention with reference to the drawings.

[0027] As Figure 1 、 Figure 3 shown, a method for calculating the electric vehicle carrying capacity of a distribution network includes the following steps: S1. Analyze the statistical quantities of electric vehicle trips to obtain the travel patterns of electric vehicle users, construct a travel chain model of electric vehicles, and describe the transfer and stay of electric vehicles between different locations through a state transition matrix.

[0028] The travel chain describes the whole process of a user's travel. Based on the travel purpose, starting from the origin, the user passes through several intermediate locations in chronological order and finally arrives at the destination. According to the national conditions of our country and the living habits of urban residents, the travel purpose areas are divided into residential areas (H areas), work and business areas (W areas), public entertainment areas (P areas), and other areas (O areas).

[0029] The expression of the state transition matrix is as follows: ; where, represents T the probability of traveling from area M to area N at time M is H , W , P or O . N is H , W , P or O . Important travel characteristics such as the daily travel frequency, the first travel time, and the travel destination of electric vehicle users can be analyzed from the travel statistics of electric vehicles. Based on this, the travel state transition matrix of electric vehicle users for 24 hours a day can be constructed.

[0030] After the electric vehicle completes a trip, the residence time at different destination nodes is fitted by a probability density function. The residence time probability of the electric vehicle in the residential area (H area) satisfies the Weibull distribution, and the probability density function is: ; where, is the shape parameter, is the scale parameter, is the residence time of the electric vehicle.

[0031] In other functional areas, the residence time probability distribution of the electric vehicle satisfies the generalized extreme value distribution. The probability density function of the residence time of the electric vehicle in W , P , O area is: ; where, is the location parameter, is the scale parameter, is the shape parameter; when , ; when , .

[0032] S2. Continuously simulate using the Monte Carlo method to obtain the charging demands of electric vehicles in each region at each time period throughout the day.

[0033] As Figure 2 shown. The specific process of continuously simulating using the Monte Carlo method to obtain the charging demands of electric vehicles in each region at each time period throughout the day is as follows: S21. Model the road network to generate a road distance matrix D, and input the road distance matrix and the road impedance matrix calculated according to the road impedance model.

[0034] Abstract the topological structure of the road network as a weighted directed graph. The topological mathematical expression model of the road network is: ;

[0035] Among them, represents the traffic road network, represents the set of all nodes in graph , represents a specific node, is the number of nodes; represents the set of all road segments in graph , represents the road segment connecting node and node ; represents the set of divided time periods, is the number of divided time periods, represents a specific time period; is the set of road segment weights, represents the weight of road segment within the time period .

[0036] The road segment weight W represents the road travel cost, and can be quantitatively studied using weights such as road length, passing speed, travel time, and travel cost. According to the urban traffic condition division standard, the saturation S evaluation index: unobstructed (0 < S < 0.6), slow (0.6 < S < 0.8), crowded (0.8 < S < 1.0), and severely congested (1.0 < S < 2.0). t At time , the impedance model of the road segment is: ;

[0037] Among them, the saturation , is the traffic flow of the road segment, C is the passing capacity; t 0 is the zero-flow travel time, are all impedance influence factors.

[0038] S22. Build an electric vehicle charging load model, initialize the basic vehicle information, extract the complete travel chains of each electric vehicle, and plan the travel path with the shortest travel time for the vehicle based on the road impedance matrix according to the travel demands of the vehicle at different times; Build an electric vehicle charging load model, including battery capacity, state of charge (SOC) of the battery, charging load, and charging duration.

[0039] The battery capacities of different types of electric vehicles follow a uniform distribution, and the expression is: ; where, q represents the battery capacity of the electric vehicle, s , h are the upper and lower limits of the battery capacity distribution range of the electric vehicle, in units of kW·h; The state of charge (SOC) of the electric vehicle battery can be used to reflect the change process of the battery power during vehicle driving. Based on the travel chain model, the SOC of the electric vehicle battery after each trip is: ; where, represents the SOC of the th electric vehicle at the end of the th trip, is the SOC at the end of the th trip, is the vehicle battery capacity; is the driving mileage of the th trip of the electric vehicle; is the power consumption per kilometer of the electric vehicle, in units of kW / km; When the electric vehicle has a charging demand, the calculation expressions for the charging load and charging duration are: ; where, represents the charging power of the th electric vehicle for target charging at t time; is a 0-1 variable representing the charging demand status of the th electric vehicle; is the target charging SOC of the electric vehicle; is the th electric vehicle's continuous charging duration starting from t time; represents the charging power.

[0040] S23. Set different numbers of electric vehicles for Monte Carlo sampling simulation, repeat the continuous simulation, analyze the charging demands corresponding to each road network node at each moment, and obtain the spatio-temporal distribution of the charging demands of electric vehicles.

[0041] S3. Construct an optimization model for the electric vehicle carrying capacity of the distribution network with the maximum access capacity.

[0042] Constructing an optimization model for the electric vehicle carrying capacity of the distribution network with the maximum access capacity includes the following steps: S31. Establish an objective function with the maximum charging satisfaction degree of both the charging station and the electric vehicle users as the goal, and establish the objective function; The specific expression of the objective function is: ; Among them, the set represents the functional area to which the EVCS belongs, . is the charging load of electric vehicles that can be accessed by the rd node in the functional area is the charging demand of the functional area predicted through simulation, is the number of nodes with charging stations accessed in the functional area , is the parameter that controls the steepness of the satisfaction function, is the parameter that affects the charging satisfaction degree of the penalty term. The operator max{X,0} means taking the larger value of the two, that is, if X≥0, the value is X, otherwise it is 0.

[0043] S32. Construct constraint conditions, including node voltage constraints, branch current constraints, branch transmission power constraints, power flow constraints, and charging demand constraints.

[0044] The expression of the node voltage constraint is: ; Among them, is the voltage of node , and are the upper and lower limits of the node voltage respectively, is the set of nodes.

[0045] The expression of the branch current constraint is: ; Among them, is the current of branch , is the upper limit of the branch current, is the set of branches.

[0046] The branch transmission power constraint is as follows: ; wherein, and are the active power and reactive power flowing through branch respectively; is the maximum apparent power flowing through branch , and its linearized form is expressed as: ; The power flow constraint adopts the Distflow model, and the expression is: ; wherein, and are the active power and reactive power at node respectively, represents the set of branches with node as the root node, represents the set of branches with node as the child node, and are the resistance and reactance of branch ef , represents the square of the node voltage amplitude, represents the square of the branch current amplitude, and are the active load and reactive load at node respectively, and are the substation active power and substation reactive power at node respectively, represents the electric vehicle charging load that can be connected at node , represents the photovoltaic output at node .

[0047] S4. Obtain the historical data sample set of photovoltaic output, construct the fuzzy set of photovoltaic output using the Wasserstein distance, establish the distributionally robust optimization model, and transform the optimization model containing the uncertainty of photovoltaic output into a deterministic electric vehicle carrying capacity model for solution.

[0048] The specific process of constructing the fuzzy set is as follows: Construct the empirical probability distribution based on the historical data of the uncertainty variable, and its specific expression is: ; wherein, is the historical empirical distribution, Denote the historical data of the uncertainty variable as a Dirac distribution centered at .

[0049] The Wasserstein distance is used to measure the distance between two distributions. The expression of the Wasserstein distance is: ; where is the true distribution, and have the support set , represents and 's joint probability distribution, represents the norm.

[0050] Given the radius<X , the data-driven Wasserstein fuzzy set is constructed as: ; where represents 's support set, constructs a ball with radius to characterize the distance between the true distribution and the historical empirical distribution .

[0051] The construction and transformation process of the distributionally robust optimization model is specifically as follows: The uncertainty of photovoltaic output is expressed as: ; where is the actual photovoltaic output at node , is the predicted value of photovoltaic output at node , is the predicted deviation value of photovoltaic output at node .

[0052] Considering that the expectation of the function with respect to the uncertainty variable is minimized under the worst distribution, the optimization problem is rewritten in the following form: ;

[0053] where is the decision variable, represents 's feasible region, is the random variable Support set.

[0054] According to the strong duality theorem, this problem can be transformed into the following solvable form: ;

[0055] wherein, The upper and lower bounds of are respectively , and are the introduced dual variables.

[0056] For ease of understanding, the method described in the present invention is simulated and tested in combination with a specific scenario. The 13-node Nguyen-Dupuis highway network in Figure 3 and the IEEE-33 node distribution network in Figure 4 are used for simulation verification. The number of electric vehicles for the traffic network simulation is set to 1200, including 1000 private cars and 200 taxis. The parameter settings of the electric vehicles are as follows: the battery capacity is uniformly distributed in the range of [35, 80] kWh; the SOC follows a normal distribution with a mean of 0.8 and a standard deviation of 0.1; the initial departure time follows a normal distribution with a mean of 7.3 and a standard deviation of 1.3. In the IEEE-33 node system, distributed photovoltaic units (PVs) with a rated capacity of 0.3 MW are connected to nodes 8, 12, 17, and 27, while electric vehicle charging stations (EVCSs) are connected to nodes 3, 5, 9, 12, 15, 20, 24, and 29.

[0057] Figure 5 The charging load distribution curves of each functional area obtained through Monte Carlo simulation are given in

[0058] . Summarizing the charging demands of each region, it can be found that the total charging load of electric vehicles during the day (9:00 - 20:00) is relatively high, overlapping with the traditional peak electricity consumption period, causing potential pressure on the operation of the distribution network. Figure 6 Based on obtaining the charging demand distribution, the analysis of the carrying capacity of electric vehicles is carried out, and the results are as shown in

[0059] . It can be seen that the charging load carrying capacities allocated to electric vehicle charging stations in different regions are different and vary at different times of the day.

[0060] Table 1 Results calculated by different methods at 12:00 ;

[0061] Therefore, by adopting the calculation method for the load-carrying capacity of electric vehicles in the distribution network described in the present invention, the problem of poor accuracy in calculating the load-carrying capacity of electric vehicles in the distribution network by the existing calculation methods can be solved, and it has the advantage of ensuring the safe and stable operation of the distribution network after connecting the charging load.

[0062] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that they can still modify or equivalently replace the technical solutions of the present invention, and these modifications or equivalent replacements cannot make the modified technical solutions deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A calculation method for the electric vehicle carrying capacity of a distribution network, characterized in that, The following steps are involved: S1. Analyze the statistics of electric vehicle travel to obtain the travel patterns of electric vehicle users, build an electric vehicle travel chain model, and describe the transfer and stay of electric vehicles between different locations through the state transition matrix; S2. Monte Carlo simulation is used to continuously simulate the charging demand of electric vehicles in various areas at various times of the day; S3. Construct an optimization model for electric vehicle carrying capacity of the distribution network with the largest access capacity; S4. Obtain a sample set of historical data on photovoltaic output, use Wasserstein distance to construct a fuzzy set of photovoltaic output, establish a distributed robust optimization model, and transform the optimization model containing photovoltaic output uncertainty into a deterministic electric vehicle carrying capacity model for solution.

2. The calculation method for the load-carrying capacity of electric vehicles in a distribution network according to claim 1, wherein: In S1, the travel chain describes the entire process of a user starting from a starting point based on the travel purpose, passing through several en route locations in chronological order, and finally arriving at the destination to complete the trip; The state transition matrix expression is: ； Among them, represents T the probability of the moment going from area M to area N . M is H , W , P or O . N is H , W , P or O . H represents the residential area, W represents the business and work area, P represents the public entertainment area, O represents other areas.

3. The method for calculating the electric vehicle carrying capacity of a distribution network according to claim 2, characterized in that: In S1, after the electric vehicle completes its journey, the time it stays at different destination nodes is fitted by a probability density function. The probability density function of the length of time the electric vehicle stays in the residential area is: ； Among them, is the shape parameter, is the scale parameter, is the parking duration of the electric vehicle; An electric vehicle in W , P , O The probability density function of the parking duration in the area is as follows: ； Among them, is a position parameter, is a scale parameter, is a shape parameter; when is the case, ; when is the case, .

4. The calculation method for the electric vehicle carrying capacity of a distribution network according to claim 3, wherein: In S2, the specific process of using the Monte Carlo simulation method to continuously simulate and obtain the charging demand of electric vehicles in various areas at various time periods throughout the day is as follows: S21. Model the road network and generate a road distance matrix D, inputting the road distance matrix and the road impedance matrix calculated according to the road impedance model; S22. Build an electric vehicle charging load model, initialize basic vehicle information, extract the complete travel chain of each electric vehicle, and plan a travel route with the shortest travel time for the vehicle based on the travel demand of the vehicle at different times according to the road impedance matrix; S23. Set different numbers of electric vehicles to perform Monte Carlo sampling simulation, repeat the simulation continuously, analyze and obtain the charging demand corresponding to each road network node at each time, and obtain the spatiotemporal distribution of the electric vehicle charging demand.

5. The calculation method for the electric vehicle carrying capacity of a distribution network according to claim 4, characterized in that: In S21, the road network topology is abstracted into a weighted directed graph, and the topological mathematical expression model of the road network is: ； Among them, represents the transportation road network, represents the graph the set of all nodes, represents a specific node, is the number of nodes; represents the graph the set of all road segments, represents the road segment connecting nodes and node ; represents the set of divided time periods, is the number of divided time periods, t represents a specific time period; is the set of road segment weights, represents the weight of road segment t within the time period ; Section weight Indicates the cost of road travel, t Time The impedance model of the section is: ； Among them, the saturation , is the traffic flow of the road section, C is the traffic capacity; t 0 is the travel time of zero flow, are all impedance influence factors.

6. The calculation method for the electric vehicle carrying capacity of a distribution network according to claim 5, characterized in that: In said S22, a charging load model of an electric vehicle is constructed, including battery capacity, battery state of charge, charging load and charging time; The battery capacity of different types of electric vehicles obeys a uniform distribution, which can be expressed as: ； Among them, represents the battery capacity of the electric vehicle, are respectively the upper and lower limits of the battery capacity distribution range of the electric vehicle, with the unit of kW·h; Based on the travel chain model, the battery state of charge (SOC) of an electric vehicle at the end of each trip is: ； Among them, represents the SOC of the th electric vehicle at the end of the th section of the journey, is the SOC at the end of the th section of the journey, is the battery capacity of the vehicle; is the mileage of the th section of the journey of the electric vehicle; is the power consumption per kilometer of the electric vehicle, with the unit of kW / km; When an electric vehicle generates a charging demand, the calculation expression of the charging load and charging time is: ； Among them, represents the charging power of the th electric vehicle for target charging at t moment; is a 0-1 variable representing the charging demand status of the th electric vehicle; is the target charging SOC of the electric vehicle; is the th electric vehicle's continuous charging duration starting from t moment; represents the charging power.

7. The calculation method for the load-carrying capacity of electric vehicles in a distribution network according to claim 6, characterized in that: In S3, constructing an optimization model for electric vehicle carrying capacity of a distribution network with the largest access capacity includes the following steps: S31, establishing an objective function; The specific expression of the objective function is: ； Among them, the set represents the functional area to which the EVCS belongs, is the electric vehicle charging load that the th node in the functional area can access, is the charging demand of the functional area predicted by simulation, is the number of nodes with charging stations access in the functional area , is a parameter that controls the steepness of the satisfaction function, is the parameter of the influence degree of the penalty term on the charging satisfaction; the operator max{X, 0} means taking the larger value of the two, that is, if X≥0, the value is X, otherwise it is 0; S32. Construct constraint conditions, which include node voltage constraints, branch current constraints, branch transmission power constraints, power flow constraints, and charging demand constraints.

8. The calculation method for the electric vehicle carrying capacity of a distribution network according to claim 7, wherein: In the S32, The expression of the node voltage constraint is: ； wherein, is the voltage of the node , and are the upper and lower limits of the node voltage respectively, is the set of nodes; The branch current constraint expression is: ； Among them, is the current of the branch , is the upper limit of the branch current, is the branch set; The branch transmission power constraint is: ； Among them, and are the active power and reactive power flowing through branch respectively; is the maximum apparent power flowing through branch Converting it into a linearized form, the expression is: ； The Distflow model is used for power flow constraints.

9. The calculation method for the electric vehicle carrying capacity of a distribution network according to claim 8, wherein: In S4, the construction process of the fuzzy set is specifically as follows: Construct an empirical probability distribution based on historical data of uncertain variables, and its specific expression is: ； Among them, is the historical empirical distribution, which represents a Dirac distribution centered on the historical data of the uncertainty variable ; Use the Wasserstein distance to measure the distance between two distributions, and the expression of the Wasserstein distance is: ； wherein, is the true distribution, and both have a support set of , denotes and 's joint probability distribution, denotes the norm; Given radius , the data-driven Wasserstein fuzzy set is constructed as follows: ； Among them, denotes the support set of A sphere with a radius of is constructed to characterize the distance between the true distribution and the historical empirical distribution There is a distance between them.

10. A calculation method for the load capacity of electric vehicles in a distribution network according to claim 9, characterized in that: In step S4, the construction and transformation process of the distributionally robust optimization model is specifically as follows: The uncertainty of photovoltaic power output is expressed as: ； Among them, is the actual PV output at node is the predicted PV output at node is the predicted PV output deviation at node Consider a function with respect to an uncertainty variable The expectation is minimized under the worst-case distribution, and the optimization problem is reformulated as follows: in the worst-case distribution ； Among them, is a decision variable, represents the feasible region of is a random variable the support set of According to the strong duality theorem, this problem is transformed into the following solvable form: ； Among them, The upper and lower bounds of are and are the introduced dual variables.

Images