阅读说明：本技术 一种基于灰狼算法的电动汽车充电站布点规划方案 (Electric vehicle charging station distribution planning scheme based on wolf algorithm ) 是由 黄文涛 邓明辉 何俊 邓长虹 王歆智 罗杰 程肖达 朱理文 于华 张博凯 余金蔓 于 2021-08-31 设计创作，主要内容包括：本发明涉及一种基于灰狼算法的电动汽车充电站布点规划方案,属电网优化规划领域。本发明通过牛顿-莱布尼兹公式,采用分段潮流对配电网动态数据进行提取,对电网和电动汽车的原始日负荷曲线进行分析比较,合理规划电动汽车充电站的容量和充放电策略,在满足其基于削峰填谷调节效益的同时,考虑电动汽车充电站作为负载和电源对电网所产生的影响,并结合交通路网,以电压偏差、输电线路裕度,网损,交通流量和充电站服务范围为优化目标,采用灰狼算法进行求解,寻得电动汽车充电站的最优布点方案。使得电动汽车充电站在缓解用户充电问题的同时,减小充电站对电网的影响,平滑负荷曲线,稳定电网的运行。(The invention relates to an electric vehicle charging station distribution planning scheme based on a wolf algorithm, and belongs to the field of power grid optimization planning. According to the method, through a Newton-Lei-Nei equation, dynamic data of the power distribution network is extracted by adopting segmented power flows, original daily load curves of a power grid and an electric automobile are analyzed and compared, the capacity and the charging and discharging strategies of the electric automobile charging station are reasonably planned, the influence of the electric automobile charging station as a load and a power supply on the power grid is considered while the regulation benefit based on peak clipping and valley filling is met, the voltage deviation, the margin of a power transmission line, the network loss, the traffic flow and the service range of the charging station are taken as optimization targets by combining a traffic road network, a gray wolf algorithm is adopted for solving, and the optimal point distribution scheme of the electric automobile charging station is found. The charging system has the advantages that the influence of the charging station on the power grid is reduced while the charging problem of a user is relieved, the load curve is smoothed, and the operation of the power grid is stabilized.)

1. An electric vehicle charging station distribution planning scheme based on a wolf algorithm is characterized in that,

step 1: acquiring power distribution network data after the electric vehicle charging station is accessed, wherein the power distribution network data comprises the size of the access capacity of the electric vehicle charging station, the load capacity of each node, the branch impedance level and the power supply level, and is used for calculating the probability load flow;

step 2: initializing access nodes of the electric vehicle charging stations in the road-electric coupling network, and initializing the positions of all wolf groups according to constraint conditions by adopting a wolf algorithm, namely, the initial positions of the M electric vehicle charging stations in the road-electric coupling network;

and step 3: calculating the probability load flow, and extracting required data according to the calculation result, wherein the method specifically comprises the following steps:

extracting power flow data of the power distribution network, wherein the actual power flow is dynamically changed, and when the time is shortened to a certain value, the actual power flow is described according to a Newton-Lebrunitz formula, and the power flow data at the moment can be approximately regarded as a deterministic power flow; segmenting the charging and discharging time periods of the electric vehicle charging station by taking the set time as an interval, and defining that the tide data is unchanged within the set time; at the moment, extracting dynamic data of the power distribution network; according to the segmented result, each segment of set time corresponds to one group of tidal current data, N groups of tidal current data are collected, all the data are integrated and averaged to obtain one group of tidal current data considering load fluctuation, the method is based on the concept of averaging, the obtained tidal current data has certain representativeness to the actual tidal current fluctuation, and the tidal current data obtained through MATLAB can be used for obtaining the target function in the step 4;

and 4, step 4: substituting the data extracted by the segmented power flow into an objective function, wherein the objective function is shown as the following formula:

S_C＝k₁S₁+k₂S₂+k₃S₃+k₄S₄+k₅S₅

S_F＝k₆S₁+k₇S₂+k₈S₃+k₉S₄+k₁₀S₅

S＝a_c×S_C+a_f×S_F

K₁+K₂+K₃+K₄+K₅＝1

K₆+K₇+K₈+K₉+K₁₀＝1

a_c+a_f＝1

S_C、S_Ftarget function for charging and discharging electric vehicle charging station at charging and discharging time, K₁、K₂……K₁₀Scaling factor in the overall target for sub-targets, a_c，a_fThe weight coefficient respectively represents the proportion of the objective function of the charging and discharging time in the total objective function, and can be adjusted according to actual needs, S₁、S₂、S₃、S₄、S₅Respectively representing bus node voltage deviation, active power margin level of an alternating current line, network loss level of the whole network, traffic flow and service range of a charging station, wherein S is an objective function obtained by considering the charging and discharging capacity of the electric vehicle charging station;

and 5: the optimal solution is solved according to the grey wolf algorithm, after the electric automobile charging station is connected to the power distribution network, the influence on the power distribution network after the charging station is connected to the power distribution network is minimum by optimizing the access place, according to the scheme provided by the scheme, M electric automobile charging stations carry out point distribution planning in the circuit-electric coupling network of N nodes, the constraint condition is complex, the calculated amount is large, therefore, the scheme provided by the scheme is solved through the grey wolf algorithm, and the grey wolf algorithm comprises:

(1) wolf cluster grading

Each solution in the gray wolf algorithm corresponds to a wolf, the leading wolf alpha represents the current optimal solution, beta and delta wolfs represent suboptimal solutions, and the other solutions are omega wolfs; alpha, beta and delta jointly determine the search direction;

(2) surrounding prey

In the wolf cluster, the orientations of α, β, δ have a great influence on the orientation of the next movement of each wolf, and in the gray wolf algorithm, the process is shown as follows:

D＝|CX_P(t)-X(t)|

X(t+1)＝X_P(t)-AD

in the formula: a and C are coefficient vectors; xp (t) is the orientation of α, β, and δ; x (t) and X (t +1) are respectively the position of any one solution before and after being influenced by xp (t);

(3) attack behavior

In the gray wolf algorithm, the traveling direction of the whole wolf group is determined by the optimal three solutions α, β, δ, and the process is embodied as an attack scheme and can be expressed as:

in the formula: x1, X2 and X3 are moving directions of the wolfs affected by alpha, beta and delta respectively, and the three determine new directions of the wolfs together;

the specific solving steps based on the gray wolf algorithm are as follows:

(1) substituting the results obtained in the steps 1 to 4 into a gray wolf algorithm for optimizing judgment, taking the first solution as an initial solution and an optimal solution, comparing and judging the initial solution and the optimal solution with the subsequent solution result, taking the optimal solution as a leading wolf, updating a and a synergistic coefficient vector A, and updating the random weight C of the influence of the current solution on the prey;

(2) according to the initial position of an electric vehicle charging stationThe results of the calculation and comparison are used to update the rank of the wolf pack and the moving direction of each wolf pack, and the formula isDetermining a new direction of the wolf pack;

(3) according to the formula D ═ CX_P(t)-X(t)|，X(t+1)＝X_P(t) -AD carrying out close surrounding on the optimal point distribution scheme of the electric vehicle charging station;

(4) according to the step 2, solving the result of each hunting in the surrounding process, keeping the optimal solution, judging whether the stopping condition is met, if so, accepting the hunting, and if not, returning to the step 2 to continue solving;

(5) and outputting an optimal point distribution planning scheme of the electric vehicle charging station until the optimal solution is obtained or the maximum convergence times are reached.

2. The grayish wolf algorithm-based electric vehicle charging station distribution planning scheme is characterized in that an electric vehicle charging station model is defined, and the Monte Carlo sampling process is utilized to obtain the number of electric vehicles which are connected to a charging station at a certain moment; the number of accessed electric vehicles at a certain time t can be approximately represented by a poisson distribution, which is defined as follows:

in the formula, λ_EV，tFor a desired value of the number of electric vehicles connected at a certain time t, n_EV，tThe number of the electric vehicles is randomly accessed; since both the expectation and variance of the poisson distribution are λ_EV，tSo the characteristic function is:

Ψ(t)＝exp{λ_EV，t(e^it-1)}

the charging load is calculated by taking 24 hours as a unit, taking every 10 minutes as an interval, and the charging load is 144 points in total in one day; the total charging load of each interval is the sum of all the charging loads of the electric automobiles:

in the formula: p_iFor the ith time period total charging power, i ═ 1,2, … …, 144: n is the total amount of the electric vehicles entering the charging station in the i time periods; p_n，iThe charging load for the nth vehicle in the ith time slot.

3. The gray wolf algorithm-based electric vehicle charging station distribution planning scheme of claim 1, wherein the load curve of the power grid is smoothed based on a charging and discharging scheme of the electric vehicle charging station; according toAnd simulating a daily charging load curve of the electric automobile through MATLAB to plan the capacity of the electric automobile charging station, wherein the Poisson distribution obeyed by the quantity of the electric automobiles and the normal distribution obeyed by the charging time of the electric automobiles comprise the following specific steps:

1) inputting parameters such as battery capacity and variance of the electric vehicles, calculating the charging load by taking 10 minutes as an interval, totaling 144 points in a day, and calculating the number N of the electric vehicles at the ith moment of the electric vehicle charging station;

2) classifying the electric vehicles at the ith moment according to the distribution of the initial charging time;

3) according to the distribution of the initial charge states of various electric automobile batteries, the initial charge states are randomly generated, and the capacity required by charging is calculated according to the formula P-C-SOC:

in the formula: c is the battery capacity of EV; p is charging power of EV;

4) returning to the second step, repeating the steps to obtain the charging load of various electric vehicles in one day;

5) superposing the charging load curves of the electric automobiles to obtain a charging load curve of a total electric automobile charging station; the peak value of the load curve is the capacity of the electric vehicle charging station.

4. The scheme as claimed in claim 1, wherein the indication of the voltage deviation of the bus nodes is defined, the operating voltage of each node cannot be exceeded for a distribution network with N bus nodes, and the bus node voltage of node i is set as U_bus-iThen, the following relation is satisfied:

U_bus-i,min≤U_bus-i≤U_bus-i,max

when the electric automobile is connected to the power distribution network, the voltage of the bus node should be close to the rated voltage U_NThe difference between them is as follows:

ΔU_bus-i＝U_bus-i-U_N

ΔU_bus-ibus node voltage U as node i_bus-iTo rated voltage U_NA deviation of (a);

with the above equation, the bus voltage deviation for node i can be shown as follows:

controlling the minimum voltage deviation of all bus nodes, taking the minimum voltage deviation as an optimization target, and defining the minimum voltage deviation as a target S₁Corresponding S₁Comprises the following steps:

5. the scheme is characterized in that an alternating current active power margin level index is defined, the alternating current active power margin level is an important index for measuring stable operation of a power system, and after the electric vehicle is connected, the transmission active power margin of each alternating current line is controlled to be the largest as possible so as to avoid the situation of power flow out of limit; the method of taking the average value has not been able to meet the actual need, nowExpressing the discrete degree of each line power flow by calculating the variance, minimizing the variance of each node as much as possible, and defining the variance as an optimization target S₂As shown in the following formula:

P_EVrepresenting the average power, P, of the AC line connecting the electric vehicle to node i_kRepresenting the active power level of the ac line connected to node i, and M representing the total number of ac lines connected to node i.

6. The scheme is characterized in that a grid loss level index of the whole grid is defined, and during the operation of the power grid, the loss of electric energy and the waste of energy can be caused by the excessive grid loss, and even the damage of a line can be caused by heating to cause an accident of a power distribution system; reducing network loss is an important economic goal; the line loss of the system is calculated by adopting a Newton-Raphson method, and the active network loss is expressed as follows:

S₃is an objective function of the grid loss of the power system, U_iAnd U_jThe voltage amplitudes of node i and node j, respectively; g_ijConductance of the branch between node i and node j; n is a radical of_LThe method comprises the steps of (1) collecting power transmission lines; theta_i，θ_jThe phase angle of the voltages at node i and node j.

7. The gray wolf algorithm-based electric vehicle charging station distribution planning scheme of claim 1, wherein a traffic flow index is defined, the layout of electric vehicle charging stations not only needs to consider the influence on the power grid, but also needs to consider the constraints of a traffic network under the actual situation, one of the main bodies served by the electric vehicle charging stations is an electric vehicle user, and therefore the planning scheme of the electric vehicle charging stations needs to capture the traffic flow to the maximum extent possible to meet the actual requirements of the user; the traffic flow objective function is as follows:

in the formula: s4 represents an objective function of traffic flow, F_i，F_jVehicle weight coefficients for the start i, end j of the route, d_{EVCS_k}Represents the length of a path k in the traffic network;

probability statistics is carried out on the charging requirements of the vehicles through the OD matrix, on the basis, simulation is carried out on the number of the vehicles to be charged of each node, and vehicle weight coefficients are as follows:

in the formula: f_iVehicle weight coefficient, n, representing node i_EVNumber of vehicles representing node, N_EVThe total number of vehicles in the traffic network is shown.

8. The scheme as claimed in claim 1, wherein a charging station service range index is defined, the layout of the charging stations maximizes the service range of the charging stations as much as possible, and the larger the service range is, the stronger the attraction of the charging stations to the electric vehicle users is, and the charging power, the electricity price and the distance between the users and the power stations of the charging stations directly influence the attraction of the charging stations to the electric vehicle users; the charging range objective function of the charging station provided by the scheme is as follows:

in the formula: s_{n_CS}The attractive force of the nth charging station to the EV user is shown, the larger the charging range of the electric vehicle charging station is, the better the charging range is, the scheme solves the planning scheme through the Hui wolf algorithm, and the S is adapted to the algorithm rule₅Representing the inverse of the charging range of an electric vehicle charging station, i.e. S₅Smaller, represents larger charge range; the attraction of the nth charging station to the EV user is as follows:

in the formula: p_{EVCS_n}Represents the charging power of the nth electric vehicle charging station, lambda represents the influence weight of other factors of the node i, and d_{EVCS_k}Indicates the length of a path k in the traffic network, E_EVPower consumption per unit distance, P_EVElectricity prices on behalf of electric vehicle charging stations;

in the formula: n is a radical of_{EVCS_n}The influence range of the nth electric vehicle charging station comprises the number of nodes; and N represents the total number of the nodes of the traffic network.

9. The scheme as claimed in claim 1, wherein the constraint condition is defined as:

node voltage constraint: u shape_imin≤U_i≤U_imaxIn the formula, U_imax、U_iminThe maximum and minimum values of the voltage at the node i;

branch capacity constraint:in the formula, P_ij、Q_ijActive and reactive on the branch; s_ijmaxIs the maximum capacity allowed;

the charging quantity and the total demand of the electric automobile are constrained: n is a radical of_EV≤N_EVCS，In the formula, N_EVNumber of charges for electric vehicle, N_EVCSThe allowed number of charging stations for the electric vehicle charging station, T is the number of charging piles in the electric vehicle charging station, S_CDZTo fill the capacity of the pile, S_EVThe SOC is the capacity of the electric automobile, and the SOC is the residual electric quantity of the electric automobile;

constraining the power balance equation:in the formula, P_i、Q_iIs the active and reactive power input at node i; p_Li、Q_LiThe active power and the reactive power of the load at the node i; g_ij、B_ijThe conductance and susceptance of the branch; u shape_i、U_jNode voltages at nodes i, j; p_DGi、Q_DGiActive power and reactive power injected into a node i by a distributed power supply; theta_ijIs the phase angle difference of the voltage;

the number of charging stations is restricted: n is more than or equal to 0_{i_EVCS}Less than or equal to 1, wherein n is_{i_EVCS}In the planning process, each road network node intelligently builds one electric vehicle charging station for the number of electric vehicle charging stations at the node i;

electric vehicle charging station service range restraint: 2 is less than or equal to N_{EVCS_i}Less than or equal to 10, wherein N is_{EVCS_i}，N_{EVCS_k}The influence ranges of the ith and kth electric vehicle charging stations respectively comprise the number of nodes;

constraint of contact ratio of an electric vehicle charging station:in the formula, N_{EVCS_i}，N_{EVCS_k}The influence ranges of the ith and kth electric vehicle charging stations respectively comprise the number of nodes, namely the service range of the electric vehicle charging station, and xi represents the same number of nodes in the service ranges of the two electric vehicle charging stations, namely the contact ratio; in the planning scheme of the scheme, the contact ratio of each electric vehicle charging station is not too high.

Technical Field

The invention relates to an electric vehicle charging station distribution planning scheme based on a wolf algorithm, in particular to a planning scheme considering the influence of a charging station as a load and a power supply on a power grid.

Background

Along with the continuous maturity of electric automobile technique, the scale that electric automobile inserts the electric wire netting is bigger and bigger, and the increase of electric automobile quantity leads to the proportion that electric automobile charging station inserts the electric wire netting also to promote constantly. The trend will certainly affect the power quality of the power grid, cause the fluctuation of the load curve of the power grid, and even affect the reliability and the economy of the operation of the power grid.

Therefore, research on electric vehicle charging stations is receiving more and more attention, and many scholars at home and abroad analyze the electric vehicle charging stations. Aiming at the problem that the stable operation of a power system is greatly influenced when a large-scale electric vehicle is connected into a power grid, the space-time distribution of the charging load of the electric vehicle in a city within one day is predicted, and then an electric vehicle charging station is built at a specific node on the basis of the prediction result; in order to reduce the influence of a large number of electric vehicles on a power grid caused by the disordered charging, an intelligent ordered charging and discharging scheme of the electric vehicles is provided, dynamic time-sharing trading of electricity prices among electric vehicle users is carried out, and guiding planning is carried out on the charging and discharging periods of the electric vehicles so as to achieve the effect of peak clipping and valley filling; the method aims at the deep interaction of a power system and a traffic system and the transfer problem of users between charging stations, and aims at considering the social annual cost of operators of electric vehicle charging stations, users of electric vehicles and a power distribution network, and the constant volume grid connection is carried out on the electric vehicle charging stations.

However, in the above researches, the energy storage characteristic of the electric vehicle charging station, that is, the discharge capacity of the power grid, is mostly not considered, the electric vehicle charging station is only used as a load to select a site, the discharge capacity of the electric vehicle charging station is not considered, and the influence of the discharge characteristic of the electric vehicle on the power grid is partially considered, but the electric price is basically adjusted to realize the ordered charging and discharging of the electric vehicle, the purpose of peak clipping and valley filling is realized through the interaction between the electric vehicle and the electric vehicle charging station, the charging and discharging capacity of the electric vehicle charging station is restricted and adjusted according to the electric price, the fluctuation is too large, the electric vehicle charging station is not considered as a distributed power supply, and the influence of the charging station as a power supply to the power grid is not considered. And aiming at the distribution planning of the electric vehicle charging station, the lowest economic cost is mostly taken as a target, and the influence on the electric vehicle charging station accessing to the power grid is rarely considered.

Therefore, if the conventional method and the objective function are used for the distribution planning of the electric vehicle charging stations, the energy storage characteristics of the electric vehicle charging stations are not fully utilized, and as the scale of the electric vehicle increases, the electric vehicle charging stations will have an increasingly large influence on the power grid, so the following three problems should be considered when considering the distribution planning problem of the electric vehicle charging stations:

1) need contrast the load curve of electric automobile and electric wire netting, carry out reasonable constant volume to the electric automobile charging station to will formulate reasonable electric automobile charging station charge-discharge scheme, come the load curve of level and smooth electric wire netting.

2) The power distribution network load flow data after the electric vehicle charging station is accessed needs to be extracted, in actual life, the power distribution network load flow data changes at any moment and is dynamic load flow, and if calculation is carried out according to traditional deterministic load flow, certain deviation can be generated between the power distribution network load flow data and an actual result. The invention needs to extract the dynamic data of the power distribution network by solving the segmented power flow.

3) The voltage deviation of the electric vehicle charging station based on the adjusting benefit to the power grid, the margin of the power transmission line, the network loss and the influence on the road network traffic flow and the service range of the charging station after the electric vehicle charging station is used as a load and a power supply to be connected into the power distribution network are fully considered, and the optimal solution is needed by means of a wolf algorithm.

Disclosure of Invention

The invention provides an electric vehicle charging station distribution planning scheme based on a wolf algorithm for the first time. Aiming at the energy storage characteristic of the electric vehicle charging station, the capacity of the electric vehicle charging station is planned by combining the load curves of the electric vehicle and the power grid, the load curve of the power grid is adjusted according to the charging and discharging scheme provided by the invention, the purpose of peak clipping and valley filling is achieved, the influence of the electric vehicle charging station serving as a load and power supply access circuit-electric coupling network on the power grid and the power grid is considered, and the optimal planning scheme is obtained through a wolf algorithm.

The above problems of the present invention are mainly solved by the following technical solutions:

an electric vehicle charging station distribution planning scheme based on a wolf algorithm is characterized in that,

step 1: collecting power distribution network data after the electric vehicle charging station is accessed, wherein the power distribution network data comprises the size of the access capacity of the electric vehicle charging station, the load capacity of each node, the branch impedance level and the power supply level, and is used for calculating the probability load flow in the step 3;

step 2: initializing access nodes of the electric vehicle charging stations in the road-electric coupling network, and initializing the positions of all wolf groups according to constraint conditions by adopting a wolf algorithm, namely, the initial positions of the M electric vehicle charging stations in the road-electric coupling network;

and step 3: calculating the probability load flow, and extracting required data according to the calculation result, wherein the method specifically comprises the following steps:

the method comprises the steps of extracting power flow data of the power distribution network, wherein the actual power flow is dynamically changed, and when the time is shortened to a certain value, the actual power flow is described according to a Newton-Lebrunitz formula, and the power flow data at the moment can be approximately regarded as a deterministic power flow. And segmenting the charging and discharging time periods of the electric vehicle charging station by taking the set time as an interval, and defining that the tide data is unchanged within the set time. And extracting the dynamic data of the power distribution network. According to the segmented result, each segment of set time corresponds to one group of tidal current data, N groups of tidal current data are collected, all the data are integrated and averaged to obtain one group of tidal current data considering load fluctuation, the method is based on the concept of averaging, the obtained tidal current data has certain representativeness to the actual tidal current fluctuation, and the tidal current data obtained through MATLAB can be used for obtaining the target function in the step 4;

and 4, step 4: substituting the data extracted by the segmented power flow into an objective function, wherein the objective function is shown as the following formula:

S_C＝k₁S₁+k₂S₂+k₃S₃+k₄S₄+k₅S₅

S_F＝k₆S₁+k₇S₂+k₈S₃+k₉S₄+k₁₀S₅

S＝a_c×S_C+a_f×S_F

K₁+K₂+K₃+K₄+K₅＝1

K₆+K₇+K₈+K₉+K₁₀＝1

a_c+a_f＝1

S_C、S_Ftarget function for charging and discharging electric vehicle charging station at charging and discharging time, K₁、K₂……K₁₀Scaling factor in the overall target for sub-targets, a_c，a_fThe weight coefficient respectively represents the proportion of the objective function of the charging and discharging time in the total objective function, and can be adjusted according to actual needs, S₁、S₂、S₃、S₄、S₅The bus node voltage deviation amount, the active power margin level of an alternating current line, the network loss level of the whole network, the traffic flow and the service range of a charging station are respectively shown, and S is an objective function obtained by considering the charging and discharging capacity of an electric vehicle charging station.

And 5: the optimal solution is solved according to the gray wolf algorithm, after the electric vehicle charging stations are connected into the power distribution network, the influence on the power distribution network after the charging stations are connected into the power distribution network is minimum by optimizing the connection places, according to the scheme provided by the scheme, M electric vehicle charging stations are subjected to point distribution planning in the circuit-electric coupling network of N nodes, the constraint condition is complex, the calculated amount is large, therefore, the scheme provided by the scheme is solved through the gray wolf algorithm, and the gray wolf algorithm is introduced as follows:

(1) wolf cluster grading

Each solution in the gray wolf algorithm corresponds to a wolf, the leading wolf alpha represents the current optimal solution, beta and delta wolfs represent suboptimal solutions, and the other solutions are omega wolfs. α, β and δ together determine the search direction.

(2) Surrounding prey

In the wolf cluster, the orientations of α, β, δ have a great influence on the orientation of the next movement of each wolf, and in the gray wolf algorithm, the process is shown as follows:

D＝|CX_P(t)-X(t)|

X(t+1)＝X_P(t)-AD

in the formula: a and C are coefficient vectors; xp (t) is the orientation of α, β, and δ; x (t) and X (t +1) are the orientations of any one solution before and after the influence of xp (t), respectively.

(3) Attack behavior

In the gray wolf algorithm, the traveling direction of the whole wolf group is determined by the optimal three solutions α, β, δ, and the process is embodied as an attack scheme and can be expressed as:

in the formula: x1, X2 and X3 are moving directions of the wolfs affected by alpha, beta and delta respectively, and the three determine new directions of the wolfs together.

The grey wolf algorithm is combined with the scheme, and the specific solving steps are as follows:

(1) substituting the results obtained in the steps 1 to 4 into a gray wolf algorithm for optimizing judgment, taking the first solution as an initial solution and an optimal solution, comparing and judging the initial solution and the optimal solution with the subsequent solution result, taking the optimal solution as a leading wolf, updating a and a synergistic coefficient vector A, and updating the random weight C of the influence of the current solution on the prey;

(2) updating the rank of the wolf groups and the moving direction of each wolf group according to the initial position calculation and comparison results of the electric vehicle charging stationDetermining a new direction of the wolf pack;

(3) according to the formula D ═ CX_P(t)-X(t)|，X(t+1)＝X_P(t) -AD carrying out close surrounding on the optimal point distribution scheme of the electric vehicle charging station;

(4) according to the step 2, solving the result of each hunting in the surrounding process, keeping the optimal solution, judging whether the stopping condition is met, if so, accepting the hunting, and if not, returning to the step 2 to continue solving;

(5) outputting an optimal point distribution planning scheme of the electric vehicle charging station until an optimal solution is obtained or the maximum convergence times are reached;

in the distribution planning scheme of the electric vehicle charging station based on the wolf algorithm, an electric vehicle charging station model is defined, and the number of electric vehicles which are connected to the charging station at a certain moment is obtained by utilizing a Monte Carlo sampling process. The number of accessed electric vehicles at a certain time t can be approximately represented by a poisson distribution, which is defined as follows:

in the formula, λ_EV,tFor a desired value of the number of electric vehicles connected at a certain time t, n_EV,tThe number of the electric vehicles is randomly accessed. Since both the expectation and variance of the poisson distribution are λ_EV,tSo the characteristic function is:

Ψ(t)＝exp{λ_EV,t(e^it-1)}

the charge load was calculated in units of 24 hours at intervals of every 10 minutes for a total of 144 points a day. The total charging load of each interval is the sum of all the charging loads of the electric automobiles:

in the formula: p_iFor the ith time period total charging power, i ═ 1,2, … …, 144: n is the total amount of the electric vehicles entering the charging station in the i time periods; p_n,iA charging load for the nth vehicle at the ith time period;

in the scheme for planning the distribution points of the electric vehicle charging station based on the wolf algorithm, a charging and discharging scheme of the electric vehicle charging station is provided to smooth a load curve of a power grid. According toAnd simulating a daily charging load curve of the electric automobile through MATLAB to plan the capacity of the electric automobile charging station, wherein the Poisson distribution obeyed by the quantity of the electric automobiles and the normal distribution obeyed by the charging time of the electric automobiles comprise the following specific steps:

1) inputting parameters such as battery capacity and variance of the electric vehicles, calculating the charging load by taking 10 minutes as an interval, totaling 144 points in a day, and calculating the number N of the electric vehicles at the ith moment of the electric vehicle charging station;

2) and classifying the electric vehicles at the ith moment according to the distribution of the initial charging time.

3) According to the distribution of the initial charge states of various electric automobile batteries, the initial charge states are randomly generated, and the capacity required by charging is calculated according to the formula P-C-SOC:

in the formula: c is the battery capacity of EV; p is charging power of EV.

4) Returning to the second step, repeating the steps to obtain the charging load of various electric vehicles in one day;

5) and superposing the charging load curves of the electric automobiles to obtain a charging load curve of the total electric automobile charging station. The peak value of the load curve is the capacity of the electric vehicle charging station.

The method comprises the following steps of carrying out research and analysis on original daily load data of a power grid in a certain area to obtain an original daily load curve of the power grid: comparing daily load curves of the electric vehicle charging station and the power grid, wherein troughs of an original daily load curve of the power grid are 2:00-7:00, 14:00-17:00, and 5:00 and 15:00 are the lowest points; the peak values are respectively 11:00-13:00 and 19:00-21:00, wherein 12:00 and 20:00 are the highest points, and after 23 points, the load curve begins to show a descending trend. The trough of the daily load curve of the electric vehicle charging station is 7:00-12:00, 18:00-21: 00; after the peak is respectively 13:00-16:00, 22: 00-next day 5:00 and 21 points, the load curve begins to show an ascending trend.

The peak power utilization time of the electric automobile is not conflicted with the power utilization peak of the power grid, even the peaks and the troughs have complementary trends, and the peak-trough difference of the load of the power grid can be reduced by utilizing the trends through an effective charging and discharging scheme of the electric automobile charging station. The specific charge-discharge scheme in the scheme is as follows: 00:00-06:00, wherein the proportion of the charging piles in the work of the electric automobile charging station is 100 percent; 06:00-10:00, wherein the proportion of the charging piles in the work of the electric automobile charging station is 50%; 10:00-13:00, wherein the proportion of charging piles in the work of the electric automobile charging station is 0% (discharging); 13:00-14:00, wherein the proportion of charging piles in the work of the electric automobile charging station is 50%; 14:00-16:00, wherein the proportion of the charging piles in the work of the electric automobile charging station is 100%; 16:00-19:00, wherein the proportion of the charging pile in the work of the electric automobile charging station is 50%; 19:00-21:30, wherein the proportion of charging piles in the work of the electric automobile charging station is 0% (discharging); 21:30-24:00, and the proportion of charging piles in the work of the electric automobile charging station is 50%.

In the distribution planning scheme of the electric vehicle charging station based on the wolf algorithm, the voltage deviation index of the bus node is defined, the running voltage of each node cannot be excessive for a power distribution network with N bus nodes, and the bus node voltage of the node i is set as U_bus-iThen, the following relation is satisfied:

U_bus-i,min≤U_bus-i≤U_bus-i,max

when the electric automobile is connected to the power distribution network, the voltage of the bus node should be close to the rated voltage U_NThe difference between them is as follows:

ΔU_bus-i＝U_bus-i-U_N

ΔU_bus-ibus node voltage U as node i_bus-iTo rated voltage U_NThe deviation of (2).

With the above equation, the bus voltage deviation for node i can be shown as follows:

controlling the minimum voltage deviation of all bus nodes, taking the minimum voltage deviation as an optimization target, and defining the minimum voltage deviation as a target S₁Corresponding S₁Comprises the following steps:

according to the gray wolf algorithm-based electric vehicle charging station arrangement planning scheme, an alternating current active power margin level index is defined, the alternating current active power margin level is an important index for measuring stable operation of a power system, and after the electric vehicle is connected, the transmission active power margin of each alternating current line is controlled to be the largest as possible, so that the situation that the power flow is out of limit is avoided. The method for taking the average value cannot meet the actual requirement, the dispersion degree of the power flow of each line is represented by the calculation of the variance, the variance of each node is minimized as much as possible, the variance is used as an optimization target, and the optimization target is defined as a target S₂As shown in the following formula:

P_EVrepresenting the average power, P, of the AC line connecting the electric vehicle to node i_kRepresenting the active power level of the ac line connected to node i, and M representing the total number of ac lines connected to node i.

In the electric vehicle charging station distribution planning scheme based on the wolf algorithm, the network loss level index of the whole network is defined, and in the operation of the power grid, the network loss is too large to cause the loss of electric energy and the waste of energy, and even the damage of a line can be caused by heating to cause an accident of a power distribution system. Reducing the grid loss is an important economic goal. The line loss of the system is calculated by adopting a Newton-Raphson method, and the active network loss is expressed as follows:

S₃is an objective function of the grid loss of the power system, U_iAnd U_jThe voltage amplitudes of node i and node j, respectively; g_ijConductance of the branch between node i and node j; n is a radical of_LThe method comprises the steps of (1) collecting power transmission lines; theta_i，θ_jThe phase angle of the voltages at node i and node j.

In the electric vehicle charging station distribution point planning scheme based on the wolf algorithm, traffic flow indexes are defined, the influence on a power grid needs to be considered in the layout of the electric vehicle charging station, the constraint of a traffic network under the actual condition needs to be considered, one of main bodies of services of the electric vehicle charging station is an electric vehicle user, and therefore the traffic flow needs to be captured to the maximum extent in the planning scheme of the electric vehicle charging station to meet the actual requirements of the user. The traffic flow objective function is as follows:

in the formula: s4 represents an objective function of traffic flow, F_i，F_jVehicle weight coefficients for the start i, end j of the route, d_{EVCS_k}Representing the length of a path k in the traffic network.

According to the scheme, probability statistics is carried out on the charging demands of the vehicles through the OD matrix, on the basis, simulation is carried out on the number of the vehicles to be charged of each node, and the vehicle weight coefficients are as follows:

in the formula: f_iVehicle weight coefficient, n, representing node i_EVNumber of vehicles representing node, N_EVThe total number of vehicles in the traffic network is shown.

In the distribution planning scheme of the electric vehicle charging station based on the wolf algorithm, the service range index of the charging station is defined, the service range of the charging station is maximized as much as possible due to the layout of the electric vehicle charging station, and the larger the service range is, the stronger the attraction of the charging station to electric vehicle users is, and the attraction of the electric vehicle charging station to the electric vehicle users is directly influenced by the charging power, the electricity price and the distance between the users and the power station of the charging station. The charging range objective function of the charging station provided by the scheme is as follows:

in the formula: s_{n_CS}The attractive force of the nth charging station to the EV user is shown, the larger the charging range of the electric vehicle charging station is, the better the charging range is, the scheme solves the planning scheme through the Hui wolf algorithm, and the S is adapted to the algorithm rule₅Representing the inverse of the charging range of an electric vehicle charging station, i.e. S₅Smaller, represents larger charge range. The attraction of the nth charging station to the EV user is as follows:

in the formula: p_{EVCS_n}Represents the charging power of the nth electric vehicle charging station, lambda represents the influence weight of other factors of the node i, and d_{EVCS_k}Indicates the length of a path k in the traffic network, E_EVPower consumption per unit distance, P_EVRepresenting the electricity price of an electric vehicle charging station.

In the formula: n is a radical of_{EVCS_n}The influence range of the nth electric vehicle charging station includes the number of nodes. And N represents the total number of the nodes of the traffic network.

In the above scheme for planning the charging station distribution of the electric vehicle based on the grayling algorithm, the constraint conditions are defined as follows:

node voltage constraint: u shape_imin≤U_i≤U_imaxIn the formula, U_imax、U_iminBeing the voltage at node iA maximum value and a minimum value.

Branch capacity constraint:in the formula, P_ij、Q_ijActive and reactive on the branch; s_ijmaxTo support the maximum capacity allowed.

The charging quantity and the total demand of the electric automobile are constrained: n is a radical of_EV≤N_EVCS，In the formula, N_EVNumber of charges for electric vehicle, N_EVCSThe allowed number of charging stations for the electric vehicle charging station, T is the number of charging piles in the electric vehicle charging station, S_CDZTo fill the capacity of the pile, S_EVThe SOC is the remaining capacity of the electric vehicle.

Constraining the power balance equation:in the formula, P_i、Q_iIs the active and reactive power input at node i; p_Li、Q_LiThe active power and the reactive power of the load at the node i; g_ij、B_ijThe conductance and susceptance of the branch; u shape_i、U_jNode voltages at nodes i, j; p_DGi、Q_DGiActive power and reactive power injected into a node i by a distributed power supply; theta_ijIs the phase angle difference of the voltage.

The number of charging stations is restricted: n is more than or equal to 0_{i_EVCS}Less than or equal to 1, wherein n is_{i_EVCS}And in the planning process, each road network node intelligently builds one electric vehicle charging station for the number of the electric vehicle charging stations at the node i.

Electric vehicle charging station service range restraint: 2 is less than or equal to N_{EVCS_i}Less than or equal to 10, wherein N is_{EVCS_i}，N_{EVCS_k}The influence ranges of the ith and kth electric vehicle charging stations respectively include the number of nodes.

Contact ratio of electric vehicle charging stationBundling:in the formula: n is a radical of_{EVCS_i}，N_{EVCS_k}The influence ranges of the ith electric vehicle charging station and the kth electric vehicle charging station respectively comprise the number of nodes, namely the service range of the electric vehicle charging station, and xi represents the same number of nodes in the service ranges of the two electric vehicle charging stations, namely the contact ratio. In the planning scheme of the scheme, the contact ratio of each electric vehicle charging station is not too high.

Therefore, the invention has the following advantages: according to the energy storage characteristic of the electric vehicle charging station, the influences of the electric vehicle charging station as a load and power access path-electric coupling network on the voltage deviation of a power grid, the margin of a power transmission line, the loss of the power grid and the traffic flow in a traffic network and the service range of the charging station are considered, the optimal solution is carried out by means of a grey wolf algorithm, and a reasonable planning scheme is finally given, so that the electric vehicle charging station has the minimum influence on the power grid while meeting the user requirements, and the load curve of the power grid can be smoothed by reasonably planning the charging and discharging scheme of the electric vehicle charging station, so that the purposes of peak clipping and valley filling are achieved, and the operation of the power grid is stabilized.

Drawings

Fig. 1 is a charging station load calculation flowchart.

Fig. 2 is a graph of the raw daily load of the grid.

FIG. 3 is a graph of raw daily load for electric vehicle charging.

FIG. 4 is a graph comparing daily load curves of a power grid and an electric vehicle.

Fig. 5 is a block diagram of an electric vehicle charging station layout plan.

Fig. 6 is a data extraction diagram of a segmented power flow.

Fig. 7 is a flow chart of power flow calculation of a distribution network including electric vehicles.

Fig. 8 is an algorithm flow chart.

Fig. 9 is a flowchart of the algorithm after the electric vehicle charging station is combined.

Fig. 10 is a diagram of a circuit-to-electrical coupling network.

FIG. 11 is a single-seat charging station targetFunction S_CScheme is compared with the figure.

FIG. 12 is a single charging station objective function S_FScheme is compared with the figure.

Figure 13 is a comparison graph of a single-seat charging station objective function S scheme.

Fig. 14 is a graph of voltage deviations of respective nodes at the time of charging at a single charging station.

Fig. 15 is a graph of power margin at each node at the time of charging at a single charging station.

Fig. 16 is a graph of voltage deviation of each node at the time of discharge of a single charging station.

Fig. 17 is a graph of power margin of each node at the time of discharge of a single charging station.

Figure 18 is a single-seat charging station solution 1 charging station service range diagram.

Figure 19 is a single charging station solution 2 charging station service range diagram.

Figure 20 is a single charging station solution 3 charging station service range diagram.

Figure 21 is a single charging station scenario 4 charging station service range diagram.

Figure 22 is a single charging station solution 5 charging station service range diagram.

Figure 23 is a single charging station solution 6 charging station service range diagram.

FIG. 24 is a multi-charging station objective function S_CScheme is compared with the figure.

FIG. 25 is a multi-charging station objective function S_FScheme is compared with the figure.

Fig. 26 is a comparison graph of a multi-charging station objective function S scheme.

Fig. 27 is a graph of voltage deviation of each node at the time of charging at a plurality of charging stations.

Fig. 28 is a graph of power margin at each node at the time of charging at a plurality of charging stations.

Fig. 29 is a graph of voltage deviation of each node at the time of discharge of a plurality of charging stations.

Fig. 30 is a graph of power margin of each node at the time of discharge of a plurality of charging stations.

Figure 31 is a multiple charging station scenario 1 charging station service area diagram.

Figure 32 is a multiple charging station scenario 2 charging station service range diagram.

Figure 33 is a multiple charging station scenario 3 charging station service range diagram.

Figure 34 is a multiple charging station scenario 4 charging station service range diagram.

Figure 35 is a multiple charging station scenario 5 charging station service range diagram.

Figure 36 is a multiple charging station scenario 6 charging station service range diagram.

Fig. 37 is a single charging station peak clipping trough map (trough).

Fig. 38 is a single charging station peak clipping and valley filling map (peak).

Fig. 39 is a graph of daily grid load after the charging station group participates in regulation.

FIG. 40 is a comparison of the process of optimizing the gray wolf algorithm and the particle swarm algorithm.

Detailed Description

The technical scheme of the invention is further specifically described by the following embodiments and the accompanying drawings.

Example (b):

the technical scheme of the invention is further specifically described in the following with reference to the attached drawings:

the key point of the charging load curve calculation is to establish a mathematical function model of the initial charging time and the initial state of charge of the electric vehicle. The total daily charging load of the electric automobile is the sum of the charging loads of all the electric automobiles connected to the power grid for charging. And (4) calculating a daily charging load curve of each vehicle according to the initial charging time and the initial charging state of each electric vehicle. And accumulating the daily charging load curves of all the electric automobiles to calculate the total charging load curve of the electric automobile charging station. The initial charge time, initial state of charge for each vehicle is shown in the following table:

TABLE 1 initial Charge time and State of Charge for various vehicles

FIG. 1 is a charging station load calculation flow chart, and as can be seen from FIG. 2, the troughs of the original daily load curve of the power grid are 2:00-7:00, 14:00-17:00, wherein 5:00 and 15:00 are the lowest points; the peak values are respectively 11:00-13:00 and 19:00-21:00, wherein 12:00 and 20:00 are the highest points, and after 23 points, the load curve begins to show a descending trend.

As can be seen from the attached figure 3, the troughs of the original daily load curve of the electric automobile are 7:00-12:00, 18:00-21: 00; after the peak is respectively 13:00-16:00, 22: 00-next day 5:00 and 21 points, the load curve begins to show an ascending trend.

As can be seen from fig. 4, the wave trough of the daily load curve of the power grid and the wave crest of the daily load curve of the electric vehicle are in the dotted area indicated by the black arrow, and the wave crest of the daily load curve of the power grid and the wave trough of the daily load curve of the electric vehicle are in the dotted area indicated by the green arrow, which indicates that the peak power utilization time of the electric vehicle is not in conflict with the power utilization peak of the power grid, even has a complementary trend, so that the peak-valley difference of the power grid load can be reduced by using the trend through an effective charging and discharging scheme of the charging station for electric vehicles.

Based on the analysis, the electric vehicle charging station is used for charging the electric vehicle in a large scale in the trough time period of the original daily load curve of the power grid, namely 00:00-6:00 and 14:00-16:00, and in the rest time periods, part of the charging piles are opened to charge the electric vehicle. In the peak time period of the original daily load curve of the power grid, namely 10:00-13:00 and 19:00-21:30, the electric vehicle charging station is discharged, so that the charging requirement of the electric vehicle is met, the load curve of the power grid can be smoothed, and the purpose of peak clipping and valley filling is achieved, and the specific proportion is shown in the following table:

TABLE 2 Charge-discharge proportional table for electric vehicle charging station

Time	Proportion (charging pile proportion)
		00:00-06:00	100％
06:00-10:00	50％
		10:00-13:00	0% (discharge)
13:00-14:00	50％
		14:00-16:00	100％
16:00-19:00	50％
		19:00-21:30	0% (discharge)
21:30-24:00	50％

In order to verify the effectiveness of the scheme, the method adopts the charging condition of the electric automobile in a typical working day to simulate. The simulation time of the charging load of the electric automobile is 00: 00-24: 00, the time interval is ten minutes, the extraction time of the segmented power flow data is 00: 00-24: 00, and the time interval is 1 minute.

The electric vehicle charging station is provided with 60 charging piles, simulation is started at an interval of ten minutes, the charge state of each electric vehicle has certain difference, so that the electric vehicle charging station can get in and get out of the station every minute, according to the table 1, the initial charging time of various vehicles obeys normal distribution, the types of the electric vehicles are unchanged in a period of time, the changed numbers (getting in and getting out of the station) are not greatly different, after a certain time, the types of the electric vehicles start to change, the charging quantity of the electric vehicles on the electric vehicle charging station under the condition that the vehicles get out of the station and get in the station at each moment is considered is obtained by simulating the operation state of the electric vehicle charging station, the sum of the electric quantity required by each electric vehicle in the charging station at each moment is the load capacity of the charging station, and the electric vehicle charging station determines the capacity through the peak value of a load curve, the battery capacity of each type of electric vehicle is shown in the following table:

TABLE 3 Battery capacity meter for various electric vehicles

Kind of electric automobile	Size of capacity
		Electric bus	55kwh
Electric taxi	45kwh
		Electric official vehicle and special vehicle	35kwh
Electric private car	35kwh

For research convenience, the invention makes the following assumptions before calculation:

1) the number of electric vehicles reaches a certain scale, and the electric vehicle charging station can be kept in an operating state all the time.

2) The invention adopts a 30-node circuit-electric coupling network, data is subject to calculation, and the electric vehicle charging station is sequentially accessed to each node in a simulation time period.

3) The types of the electric automobiles obey Poisson distribution, the quantity of the electric automobiles obeys normal distribution, the capacity of the electric automobile charging station is 1.3MW according to the result of daily load curve simulation of the electric automobiles, the expected charge state of an owner is 1, and the charging pile efficiency eta is 0.95.

4) The conventional charging power of the electric automobile is 2-4 KW, the quick charging power is 32-64 KW, the charging process is simplified to be a constant power characteristic, the conventional charging power is 3KW, and the quick charging power is 48 KW.

A gray wolf algorithm-based electric vehicle charging station distribution planning scheme considers the adjustment benefits based on peak clipping and valley filling and the influence on a power grid, and adopts the gray wolf algorithm to solve the optimal scheme. The mathematical model is as follows:

1. objective function 1:

in the objective function, S₁Indicating the amount of bus node voltage deviation, U_bus-iRepresenting bus node voltage, U_NRepresenting the rated voltage and N representing the number of bus nodes.

The objective function 2:

in the objective function, S₂Representing the active power margin, P, of the alternating current_EVRepresenting the average power, P, of the AC line connected to node i_kRepresenting the active power level of the ac line connected to node i, and M representing the total number of ac lines connected to node i.

The objective function 3:

in the objective function, S₃Is an objective function of the grid loss of the power system, U_iAnd U_jAre respectively node i and nodeThe voltage amplitude of j; g_ijConductance of the branch between node i and node j; n is a radical of_LThe method comprises the steps of (1) collecting power transmission lines; theta_i，θ_jThe phase angle of the voltages at node i and node j.

The objective function 4:

in the objective function, S₄An objective function representing the traffic flow, F_i，F_jVehicle weight coefficients for the start i, end j of the route, d_{EVCS_k}Representing the length of a path k in the traffic network.

The objective function 5:

in the objective function, S_{n_CS}The attractive force of the nth charging station to the EV user is shown, the larger the charging range of the electric vehicle charging station is, the better the charging range is, the scheme solves the planning scheme through the Hui wolf algorithm, and the S is adapted to the algorithm rule₅Representing the inverse of the charging range of an electric vehicle charging station, i.e. S₅Smaller, represents larger charge range.

According to the working period of the electric vehicle charging station, the influence of the electric vehicle charging station on the power grid is considered in two situations, and the objective function is shown as the following formula:

S_C＝k₁S₁+k₂S₂+k₃S₃+k₄S₄+k₅S₅

S_F＝k₆S₁+k₇S₂+k₈S₃+k₉S₄+k₁₀S₅

S＝a_c×S_C+a_f×S_F

K₁+K₂+K₃+K₄+K₅＝1

K₆+K₇+K₈+K₉+K₁₀＝1

a_c+a_f＝1

S_C、S_Ftarget function for charging and discharging electric vehicle charging station at charging and discharging time, K₁、K₂……K₁₀Scaling factor in the overall target for sub-targets, a_c，a_fThe weight coefficient respectively represents the proportion of the objective function of the charging and discharging time in the total objective function, and can be adjusted according to actual needs, S₁、S₂、S₃、S₄、S₅The bus node voltage deviation amount, the active power margin level of an alternating current line, the network loss level of the whole network, the traffic flow and the service range of a charging station are respectively shown, and S is an objective function obtained by considering the charging and discharging capacity of an electric vehicle charging station.

2. Constraint conditions are as follows:

1) node voltage constraint

U_imin≤U_i≤U_imax

In the formula: u shape_imax、U_iminThe maximum and minimum values of the voltage at node i.

2) Branch capacity constraint

In the formula: p_ij、Q_ijActive and reactive on the branch;

S_ijmaxto support the maximum capacity allowed.

3) Electric automobile charging quantity and total demand constraint

N_EV≤N_EVCS

In the formula: n is a radical of_EVNumber of charges for electric vehicle, N_EVCSThe allowed number of charging stations for the electric vehicle charging station, T is the number of charging piles in the electric vehicle charging station, S_CDZTo fill the capacity of the pile, S_EVThe SOC is the remaining capacity of the electric vehicle.

4) Constraining the power balance equation:

in the formula: p_i、Q_iIs the active and reactive power input at node i;

P_Li、Q_Lithe active power and the reactive power of the load at the node i;

G_ij、B_ijthe conductance and susceptance of the branch;

U_i、U_jnode voltages at nodes i, j;

P_DGi、Q_DGiactive power and reactive power injected into a node i by a distributed power supply;

θ_ijis the phase angle difference of voltage

5) Charging station quantity constraints

0≤n_{i_EVCS}≤1

In the formula: n is_{i_EVCS}And in the planning process, each road network node intelligently builds one electric vehicle charging station for the number of the electric vehicle charging stations at the node i.

6) Electric vehicle charging station service range constraints

2≤N_{EVCS_i}≤10

In the formula: n is a radical of_{EVCS_i}，N_{EVCS_k}The influence ranges of the ith and kth electric vehicle charging stations respectively include the number of nodes.

7) Contact ratio constraint of electric vehicle charging station

In the formula: n is a radical of_{EVCS_i}，N_{EVCS_k}The influence ranges of the ith and the kth electric vehicle charging stations respectively comprise the number of nodesI.e. the service range of the electric vehicle charging station, ξ represents the same number of nodes, i.e. the degree of coincidence, in the service ranges of the two electric vehicle charging stations. In the planning scheme of the scheme, the contact ratio of each electric vehicle charging station is not too high.

3. Solving the mathematical model by using a gray wolf optimization algorithm, as shown in fig. 9, specifically comprises the following steps:

step 1: inputting the load quantity of each node, the branch impedance level, the power supply level and the like;

step 2: initializing the capacity of the electric vehicle charging stations and the positions of all wolf groups according to the model parameters and the constraint conditions, namely, the initial positions of the M electric vehicle charging stations;

and step 3: calculating a target function value by a power distribution network dynamic data extraction method provided by the third chapter according to the position of the initialized wolf group;

and 4, step 4: carrying out optimization judgment according to the calculation result, taking the optimal solution as a leading wolf, updating a and a synergistic coefficient vector A, and updating the random weight C of the influence of the current solution on a prey;

and 5: updating the rank of the wolf groups and the moving direction of each wolf group according to the initial position calculation and comparison results of the electric vehicle charging stationDetermining a new direction of the wolf pack;

step 6: according to the formula D ═ CX_P(t)-X(t)|，X(t+1)＝X_P(t) -AD close-in surrounding an optimal distribution point of the electric vehicle charging station;

and 7: according to the step 2, solving the result of each hunting in the surrounding process, keeping the optimal solution, judging whether the stopping condition is met, if so, accepting the hunting, and if not, returning to the step 2 to continue solving;

and 8: outputting an optimal point distribution planning scheme of the electric vehicle charging station until an optimal solution is obtained or the maximum convergence times are reached;

4. in order to verify the beneficial effect of the method of the invention, the following simulation experiment is carried out:

according to the scheme, simulation analysis is performed by adopting a distribution network of IEEE 30 and coupling frames of 30 road network nodes, the corresponding distribution network-traffic network coupling frame is shown in figure 10, the unit distance between the nodes in the figure is 1KM, the positions of corresponding charging stations are represented by grid filling, and in figure 10, the number of the charging stations takes M as an example and 3 is taken as an example. And the influence factor weight of each node is shown in the following table:

TABLE 4 influence factor weight Table

Node point	Influencing factor	Node point	Influencing factor	Node point	Influencing factor
						1	1.3	11	0.8	21	0.9
2	1.4	12	1.0	22	0.7
						3	1.5	13	1.2	23	1.4
4	1.3	14	1.3	24	1.2
						5	0.7	15	1.3	25	1.1
6	0.9	16	1.1	26	1.0
						7	0.7	17	0.9	27	0.9
8	0.6	18	0.7	28	0.7
						9	1.2	19	1.2	29	0.6
10	1.1	20	1.2	30	0.2

1) Simulation of single charging station planning scheme

In the IEEE-30 node standard calculation example, 1,2,5,8,11,13 are reference nodes and PV nodes, generally power plants and power stations, and therefore are not considered as nodes for electric vehicle charging station distribution. When M is equal to 1, the gray wolf algorithm is adopted for optimization solution, in the solution process, ten times of circulation is taken as one week, the last solution result of each week is taken, the optimization process of the gray wolf algorithm is embodied through data extraction, and the extraction results are shown in the following table:

table 5 single charging station scheme table

Scheme(s)	Node point
		Scheme(s)1	7
Scheme 2	12
		Scheme 3	19
Scheme 4	24
		Scheme 5	30
Scheme 6	17

When a single electric vehicle charging station accesses a circuit-electric coupling network according to the planning scheme provided by the scheme, the traffic flow and the service range of each scheme are shown in the following table through a gray wolf algorithm:

table 6 road network factor table

Scheme(s)	Nodal traffic flow	S5Value of objective function	Service scope
				Scheme 1	110.875	10.564	3 nodes
Scheme 2	34.001	8.231	4 nodes
				Scheme 3	24.298	5.215	6 nodes
Scheme 4	22.700	1.061	10 nodes
				Scheme 5	8.3467	4.103	7 nodes
Scheme 6	8.3328	3.213	9 nodes

The traffic flow objective function value is set to satisfy the optimization of the gray wolf algorithm optimization algorithm, and is set in the form of reciprocal, that is, the smaller the objective function value, the larger the traffic flow of the node is, so the traffic flow of the scheme 6 is better than the flow values of other schemes as can be seen from the above table. The number of service range nodes in scheme 4 is N or less according to equation 2_{EVCS_i}The constraint less than or equal to 10 is selected as 10 service nodes, and the service of each scheme is specifiedThe range nodes are illustrated in fig. 18-23.

And obtaining the target function S through the gray wolf algorithm_CThe data comparison of each scheme is as shown in figure 11:

as can be seen from fig. 11, the objective function value of scheme 6 is optimal compared to the other schemes when the single-seat electric vehicle charging station is used as a load access path-electric coupling network. However, the planning scheme provided by the scheme not only needs to consider the load characteristic of the electric vehicle charging station, but also needs to consider the energy storage characteristic of the electric vehicle charging station, so that when the single-seat electric vehicle charging station is used as a power access circuit-electric coupling network, the target function S is obtained through the Husky algorithm_FThe data comparison of each scheme is as shown in figure 12:

as can be seen from fig. 12, the objective function values of the solutions 5 and 6 are significantly better than those of the other four solutions, which also represents the optimization process and effectiveness of the grayling algorithm, the solution 5 is slightly larger than the solution 6 in service range, but the voltage deviation and the power margin of the solution 6 are significantly better than those of the solution 5, and in combination with the table 6, fig. 18 to 23, although the service range of the electric vehicle charging station in the solution 5 is larger than that of the solution 6, the position of the electric vehicle charging station is too close to the end of the coupling network, and the geographical position is not as reasonable as the solution 6.

Incorporating an objective function S_CAnd S_FA according to the formula_c×S_C+a_f×S_FWhen a single electric vehicle charging station is connected to the power distribution network, optimization data of 6 schemes are obtained, and the corresponding optimization results are shown in the attached drawing 13:

by combining table 6 and the comparison of the above schemes, it can be seen that the objective function value corresponding to scheme 6 is optimal, and therefore, the index corresponding to the connection of a single-seat electric vehicle charging station to node number 17 is better. When the electric vehicle charging station is connected to a power distribution network as a load, the voltage deviation of each node in 24 hours is shown in the attached figure 14:

the power margins at each node at 24 hours are shown in fig. 15:

when the electric vehicle charging station is connected to a power distribution network as a power supply, the voltage deviation of each node in 24 hours is shown in the attached figure 16:

the power margins at each node at 24 hours are shown in fig. 17:

the service scope of the electric vehicle charging station of the schemes 1-6 in the road network is shown in the attached figures 18-23.

According to the simulation result, according to the optimization scheme provided by the scheme, when a single electric vehicle charging station is connected into a circuit-electric coupling network, the charging requirement can be met, the service range is maximized, the influence of the electric vehicle charging station on a power grid can be effectively reduced, and the operation of the power grid is stabilized.

2) Simulation of multiple charging station planning schemes

When M is 3, the point distribution planning is carried out on the road-electric coupling network according to the scheme provided by the scheme, and the solution scheme in the optimization process of the gray wolf algorithm is sampled and extracted as described in the simulation of the planning scheme of the single charging station, wherein the specific scheme is shown in the following table:

table 7 scheme table with multiple charging stations

Scheme(s)	Node point
		Scheme 1	17、7、28
Scheme 2	22、21、29
		Scheme 3	22、13、23
Scheme 4	20、7、13
		Scheme 5	30、15、11
Scheme 6	17、21、18

Combining a traffic network, solving by a wolf algorithm, and when the electric vehicle charging station is accessed according to the scheme, the traffic flow and the service range of each scheme are shown as the following table:

table 8 road network factor table

As shown in the table, the traffic flow of the node where the electric vehicle charging station is located in the scheme 6 is superior to that in other schemes, and the schemes 3-5 are that N is more than or equal to N according to the formula 2_{EVCS_i}The constraint of less than or equal to 10, the minimum is 2 nodes, and the maximum is 10 nodes, and the service range nodes of the specific schemes are illustrated in the attached figures 31-36.

When the electric vehicle charging station is used as a load access circuit-electric coupling network, the optimization data comparison of each scheme is as shown in the accompanying figure 24:

as shown in fig. 24, the objective function of the scheme 6 is obviously better than the other five schemes, and in addition, the power characteristics of the electric vehicle charging stations are considered, when a plurality of electric vehicle charging stations are used as a power access circuit-electric coupling network, the objective function S of each scheme is obtained through the gray wolf algorithm_FThe optimization data is shown in figure 25:

as shown in fig. 25, the respective objective functions of the solution 6 are also optimized when the electric vehicle charging station is connected as a power supply to the coupling network.

Incorporating an objective function S_CAnd S_FA according to the formula_c×S_C+a_f×S_FWhen a plurality of electric vehicle charging stations access the circuit-electric coupling network, optimization data of 6 schemes are obtained, and the corresponding optimization result is shown in figure 26：

By comparing the above schemes with table 8, it can be seen that the objective function value according to scheme 6 is optimal. When the electric vehicle charging station is connected to a power distribution network as a load, the voltage deviation of each node in 24 hours is shown in the attached figure 27:

the power margins at each node at 24 hours are shown in fig. 28:

when the electric vehicle charging station is connected to a power distribution network as a power supply, the voltage deviation of each node in 24 hours is shown in the attached figure 29:

the power margins at each node at 24 hours are shown in fig. 30:

the service ranges of the electric vehicle charging stations of the schemes 1-5 in the road network are shown in the accompanying drawings 31-36:

the service scope information table for each scheme is shown in the following table:

table 9 service scope information table

Scheme(s)	Coverage rate of range	Number of non-covered nodes	Ratio of overlap	Percentage of non-coverage
					Scheme 1	50.00％	15	16.67％	50.00％
Scheme 2	36.67％	19	23.33％	63.33％
					Scheme 3	50.00％	15	10.00％	50.00％
Scheme 4	73.33％	8	6.67％	26.67％
					Scheme 5	73.33％	8	0％	26.67％
Scheme 6	83.33％	5	13.33％	16.67％

As can be seen from the above table, the coverage area of the scheme 6 is the largest, the number of uncovered nodes of the scheme 6 is the smallest in the whole road-electric coupling network, although the coincidence rate of the service area is slightly higher than that of the schemes 3-5, by comparing fig. 31-36, the number of uncovered nodes of the scheme 3 is too large to meet the requirement of the EV user, the coincidence rate of the service areas of the scheme 4 and the scheme 5 is superior, but the position of the electric vehicle charging station is poorer than that of the scheme 6, the position of the electric vehicle charging station in the scheme 4 is slightly shifted to the front section of the road-electric coupling network, the position of the electric vehicle charging station in the scheme 5 is shifted to the end of the network, and the position in the scheme 6 can meet the requirement of the EV user to the maximum, to sum up, the scheme 6 is the optimal scheme compared with other schemes.

3) Regulation benefit simulation based on peak clipping and valley filling

After the electric vehicle charging station is connected to the circuit-electric coupling network according to the optimization scheme provided by the scheme, the load peak can be effectively reduced, the load valley is filled, and the daily load curve of the power grid after the single electric vehicle charging station participates in regulation is shown in the attached figures 37-38:

in the attached figures 37 to 38, a daily load curve of a power grid part is intercepted, and after the daily load curve of the power grid is adjusted by the electric vehicle charging station according to the scheme provided by the scheme, an obvious peak clipping and valley filling trend can be seen.

Because of the restriction of capacity, the regulation of this scheme electric automobile charging station to the distribution network load curve is not obvious, nevertheless according to the optimization scheme that this scheme proposed, after the station number of electric automobile charging station reached certain scale, electric automobile charging station crowd participated in the electric wire netting daily load curve after the regulation as shown in fig. 39:

as shown in fig. 39, the fluctuation of the daily load curve is obviously suppressed, the electric vehicle charging station discharges to the power grid to reduce the load peak of the power grid, and the electric vehicle charging station charges to the EV to improve the load valley of the power grid, so that the operation of the power grid is stabilized. Therefore, after the electric vehicle charging station is optimally operated according to the charging and discharging scheme provided by the scheme, the daily load curve of the power grid can be effectively adjusted, the purpose of peak clipping and valley filling is achieved, the power supply reliability is improved, and the validity of the electric vehicle charging station planning scheme researched by the scheme is fully proved.

4) Comparison of optimization algorithms

The scheme compares the optimizing process and the result of the gray wolf algorithm and the PSO algorithm, and the comparison result is shown as the attached drawing 40:

as shown in fig. 40, the grey wolf algorithm is superior to the PSO algorithm in terms of convergence speed and solution accuracy, and compared with the PSO algorithm, the grey wolf algorithm has a higher solution speed and a smaller occupied memory while satisfying the solution accuracy, which also reflects the reliability of the optimal placement planning scheme for the electric vehicle charging station based on the grey wolf algorithm.

The specific embodiments described herein are merely illustrative of the spirit of the invention. Various modifications or additions may be made to the described embodiments or alternatives may be employed by those skilled in the art without departing from the spirit or ambit of the invention as defined in the appended claims.