阅读说明：本技术 一种基于双层优化算法的电动汽车充电站有序充电方法 (electric vehicle charging station ordered charging method based on double-layer optimization algorithm ) 是由 沈华 林林 陈葛亮 殷俊 李勇生 金震 胡楠 傅慧华 张丁溶 于 2019-11-20 设计创作，主要内容包括：本发明提供了一种基于双层优化算法的电动汽车充电站有序充电方法,上层采用遗传算法对各充电站调度计划P<Sub>j,t</Sub>编码形成初始种群,适应度函数为F,然后通过算法迭代最终得到P<Sub>j,t</Sub>最优解；然后将最优解传递给下层的各充电站；下层考虑电动汽车充电路径选择机制,采用蚁群算法求解；首先输入电动汽车参数、位置信息和充电站信息等,计算下层模型最小调度偏差函数f<Sub>1</Sub>,并基于最小偏差及约束,采用蚁群算法计算目标函数f,形成电动汽车最佳充电方案,并返回上层继续优化,求解终止条件为上下层调度偏差满足收敛判据。本发明上层模型优化目标为实现负荷削峰填谷,下层模型用以减少电动汽车充电出行时间和充电费用。(The invention provides electric vehicle charging station ordered charging methods based on double-layer optimization algorithm, wherein the upper layer adopts genetic algorithm to schedule plan P for each charging station j,t Encoding to form an initial population, performing fitness function F, and performing algorithm iteration to finally obtain P j,t An optimal solution; then transmitting the optimal solution to each charging station on the lower layer; the lower layer considers a charging path selection mechanism of the electric automobile and adopts an ant colony algorithm to solve; firstly, inputting parameters, position information, charging station information and the like of the electric vehicle, and calculating a minimum dispatching deviation function f of a lower layer model 1 And based on the minimum deviation and the constraint, adopting an ant colony algorithm to calculate a target function f to form an optimal charging scheme of the electric automobile, returning to the upper layer for continuous optimization, and solving a termination condition to ensure that the upper and lower layer scheduling deviations meet a convergence criterion. Book (I)The upper layer model is optimized to realize load peak clipping and valley filling, and the lower layer model is used for reducing the charging travel time and charging cost of the electric automobile.)

1, electric vehicle charging station ordered charging method based on double-layer optimization algorithm, characterized in that, the electric vehicle charging station ordered charging method specifically includes the following steps:

step (1), determining an ordered charging control framework of an electric vehicle charging station, comprising three layers of structures: the system comprises a power distribution network management layer, a charging station operation layer and an electric vehicle user layer;

step (2), an upper and lower layer scheduling model is established based on the ordered charging control framework of the electric vehicle charging stations, an upper layer power distribution network management layer formulates a scheduling plan of each charging station, and a lower layer charging station operation layer determines a specific charging scheme of each electric vehicle according to the upper layer scheduling plan;

optimizing an upper model objective function F by adopting a genetic algorithm;

step (4), the obtained scheduling plan P_jTransmitting the data to a lower charging pile;

step (5), initializing an electric automobile serial number i, and enabling i to be 0;

step (6), inputting electric vehicle parameters, position information and charging station information;

step (7), calculating the minimum scheduling deviation function f of the lower layer model₁；

Step (8), calculating a target function by adopting an ant colony algorithm with the minimum deviation and a convergence criterion as constraints;

step (9), obtaining an optimal charging station j and generating an optimal charging path;

step (10), judging whether I is smaller than the number I of electric vehicles in the charging station j, if so, entering step (11), otherwise, I is I +1, and returning to step (6);

step (11), obtaining a charging load scheduling plan of a lower model charging station j;

step (12), judging whether the convergence criterion is met:

and (13) outputting the optimal scheduling plan of each charging station and the optimal charging scheme of each electric vehicle.

2. The electric vehicle charging station ordered charging method based on double-layer optimization algorithm, wherein the charging station provides three charging modes for users, which are:

fast dc charging mode: the battery of the electric automobile is directly charged in a large current mode through the off-board charger, about 2 hours are needed for fully charging the battery which is deeply discharged in the mode, and the theoretical maximum time can be reduced to 0.5 hour;

slow ac charging mode: compared with the quick direct-current charging, the slow charging power is lower and is about 6.6kW, the charging current is only 16A, and the charging time is prolonged and is 6-8 h;

battery replacement mode: when the energy of the battery of the electric automobile is low, the battery pack with low electric quantity is replaced by the battery pack with full electric quantity, so that the aim of quickly supplementing the electric energy of the electric automobile is fulfilled; the battery replacement operation is completed within 3min, so that the driving range of the electric automobile is infinitely extended.

3. The electric vehicle charging station ordered charging method based on double-layer optimization algorithm, according to claim 2, wherein the upper and lower layer scheduling models in step (2) are specifically:

(a) upper optimization model

The upper layer model needs to consider that the fluctuation of the charging load is unfavorable for the operation of the power grid, the comprehensive minimum of the deviation of the dispatching plan is determined by the load variance of the power distribution network and the upper and lower layer models as an optimization target, and the objective function is as follows:

where T is the total number of scheduling periods, P_t ^aThe conventional residential electricity load is a time period t; p_j,tThe charging load of a charging station J is a time period t, and J is the number of the charging stations; p^avα is a weighting coefficient which represents the punishment to the deviation of the upper and lower layers of dispatching plans;

node voltage amplitude constraints

U_m ^min≤U_m≤U_m ^max(2)

In the formula of U_m ^maxAnd U_m ^minThe upper and lower voltage limits of the node m are respectively;

line transmission power constraint

|P_l,t|≤P_l ^max(3)

In the formula P_l ^maxMaximum operating transmission power for line l;

(b) lower optimization model

The lower layer optimization is based on the premise that the deviation between the scheduling plan determined by the upper layer and the lower layer response charging load is minimum, and the objective function is as follows:

in the formula

the lower layer model calculates a response charging plan through the formula (5) and feeds the response charging plan back to the upper layer, and the upper layer carries out optimization calculation again until the convergence criterion is met:

in the formula, ε is convergence accuracy, and is taken as 10^-3；

Maximum driving mileage constraint of electric automobile

In the formula, E_i ^BI, the rated capacity of a battery of the electric automobile; SOC_i,tThe SOC of the electric vehicle i is a time period t;the power consumption of the electric automobile is per hundred miles;

electric vehicle SOC constraint

SOC_i ^min≤SOC_i,t≤SOC_i ^max(8)

In the formula, SOC_i ^max、SOC_i ^minRespectively being electric steamThe upper and lower SOC limits of the vehicle are 1 and 0.1 respectively.

Technical Field

The invention belongs to the technical field of electric automobiles, and particularly relates to an electric automobile charging station ordered charging method based on a double-layer optimization algorithm.

Background

In recent years, global energy crisis is increasingly prominent, air pollution and global temperature rise are increasingly harmed, and people pay more attention to energy and environmental problems.

Because the electric automobile load has the characteristic of space-time randomness, the large-scale charging load of the electric automobile can generate adverse effects of increasing the peak-valley difference of a power grid, disturbing the stable operation of the power distribution network and the like. How to reduce the impact of the charging load of the electric vehicle on the power grid so as to be beneficial to the orderly charging of the safe and economic operation of the power grid as a target becomes a hot spot of the current research. The charging method of the present research mostly assumes that the electric vehicle adopts a slow or fast charging mode, and a charging station in the future can provide a plurality of charging modes such as slow charging, fast charging and battery replacement, etc., and make a dynamic time-sharing charging electricity price to guide the electric vehicle user to select according to the self-demand, but there is a few literature designs so far.

Disclosure of Invention

In order to overcome the defects of the prior art, the invention provides electric vehicle charging station ordered charging methods based on double-layer optimization models, wherein the upper-layer model optimization aims at realizing load peak clipping and valley filling, and the lower-layer model is used for reducing the charging trip time and charging cost of an electric vehicle.

The invention specifically relates to an electric vehicle charging station ordered charging method based on a double-layer optimization algorithm, which comprises the following steps:

step (1), determining an ordered charging control framework of an electric vehicle charging station, comprising three layers of structures: the system comprises a power distribution network management layer, a charging station operation layer and an electric vehicle user layer;

step (2), an upper and lower layer scheduling model is established based on the ordered charging control framework of the electric vehicle charging stations, an upper layer power distribution network management layer formulates a scheduling plan of each charging station, and a lower layer charging station operation layer determines a specific charging scheme of each electric vehicle according to the upper layer scheduling plan;

optimizing an upper model objective function F by adopting a genetic algorithm;

step (4), the obtained scheduling plan P_jTransmitting the data to a lower charging pile;

step (5), initializing an electric automobile serial number i, and enabling i to be 0;

step (6), inputting electric vehicle parameters, position information and charging station information;

step (7), calculating the minimum scheduling deviation function f of the lower layer model₁；

Step (8), calculating a target function by adopting an ant colony algorithm with the minimum deviation and a convergence criterion as constraints;

step (9), obtaining an optimal charging station j and generating an optimal charging path;

step (10), judging whether I is smaller than the number I of electric vehicles in the charging station j, if so, entering step (11), otherwise, I is I +1, and returning to step (6);

step (11), obtaining a charging load scheduling plan of a lower model charging station j;

step (12), judging whether the convergence criterion is met:

if yes, entering the step (13), if not, transferring the charging load responded by the lower layer to the upper layer for continuous optimization, and returning to the step (1); wherein, P_j,tThe charging load of station j is charged for time period t,

charging corresponding loads of a charging station j at a time interval t obtained by optimization of the lower layer;

and (13) outputting the optimal scheduling plan of each charging station and the optimal charging scheme of each electric vehicle.

, the charging station provides the user with three charging modes, which are:

fast dc charging mode: the battery of the electric automobile is directly charged in a large current mode through the off-board charger, about 2 hours are needed for fully charging the battery which is deeply discharged in the mode, and the theoretical maximum time can be reduced to 0.5 hour;

slow ac charging mode: compared with the quick direct-current charging, the slow charging power is lower and is about 6.6kW, the charging current is only 16A, and the charging time is prolonged and is 6-8 h;

battery replacement mode: when the energy of the battery of the electric automobile is low, the battery pack with low electric quantity is replaced by the battery pack with full electric quantity, so that the aim of quickly supplementing the electric energy of the electric automobile is fulfilled; the battery replacement operation is completed within 3min, so that the driving range of the electric automobile is infinitely extended.

, the upper and lower layer scheduling models in the step (2) are specifically:

(a) upper optimization model

The upper layer model needs to consider that the fluctuation of the charging load is unfavorable for the operation of the power grid, the comprehensive minimum of the deviation of the dispatching plan is determined by the load variance of the power distribution network and the upper and lower layer models as an optimization target, and the objective function is as follows:

where T is the total number of scheduling periods, P_t ^aThe conventional residential electricity load is a time period t; p_j,tThe charging load of a charging station J is a time period t, and J is the number of the charging stations; p^avα is a weighting coefficient which represents the punishment to the deviation of the upper and lower layers of dispatching plans;

charging corresponding loads of a charging station j at a time interval t obtained by optimization of the lower layer;

node voltage amplitude constraints

U_m ^min≤U_m≤U_m ^max(2)

In the formula of U_m ^maxAnd U_m ^minThe upper and lower voltage limits of the node m are respectively;

line transmission power constraint

|P_l,t|≤P_l ^max(3)

In the formula P_l ^maxMaximum operating transmission power for line l;

(b) lower optimization model

The lower layer optimization is based on the premise that the deviation between the scheduling plan determined by the upper layer and the lower layer response charging load is minimum, and the objective function is as follows:

in the formula

Responding to the charging plan for the lower layer; i is the number of electric vehicles in the charging station j; x is the number of_i,k,tA decision variable is adopted, wherein the value equal to 1 represents that the electric automobile i is in a charging state in the mode k in the time period t, and the value equal to 0 represents that the electric automobile i is not in the charging state; wherein k is 1 for fast charging mode, k is 2 for slow charging mode, and k is 3 for battery changing mode;

the lower layer model calculates a response charging plan through the formula (5) and feeds the response charging plan back to the upper layer, and the upper layer carries out optimization calculation again until the convergence criterion is met:

in the formula, ε is convergence accuracy, and is taken as 10^-3；

Maximum driving mileage constraint of electric automobile

In the formula, E_i ^BI, the rated capacity of a battery of the electric automobile; SOC_i,tThe SOC of the electric vehicle i is a time period t;

the power consumption of the electric automobile is per hundred miles;

electric vehicle SOC constraint

SOC_i ^min≤SOC_i,t≤SOC_i ^max(8)

In the formula, SOC_i ^max、SOC_i ^minRespectively taking 1 and 0.1 as the upper and lower SOC limits of the electric automobile.

The invention has the beneficial effects that:

(1) the charging station is considered to provide a fast direct current charging mode, a slow alternating current charging mode and a battery replacement mode for a user, and the three charging modes expand the requirements of the user and are beneficial to the trip of the electric automobile of the user;

(2) an upper-layer power distribution network management layer and a lower-layer power distribution network management layer are adopted to establish a scheduling plan of each charging station, and a lower-layer charging station operation layer determines a specific charging scheme of each electric vehicle according to the upper-layer scheduling plan, so that the optimal charging path selection of the electric vehicle is facilitated;

(3) compared with disordered charging, the optimized ordered charging method reduces the load of the charging station in the peak period of the system load, and transfers part of the charging load to the electricity consumption valley period, so as to play a role of 'peak clipping and valley filling', and ensure that the time-space distribution of the load of the power distribution network is more reasonable;

(4) under the ordered control, the charging load is obviously adjusted, and the control factor of the minimum load variance is considered, so that the fluctuation of the load is obviously reduced;

(5) the optimal scheduling of the electric automobile charging through the lower model can reduce the travel time and charging cost of the electric automobile charging.

Drawings

FIG. 1 is an ordered charging control framework for an electric vehicle charging station according to the present invention;

FIG. 2 is a flow chart of an electric vehicle ordered charging method based on a double-layer optimization model.

Detailed Description

The following describes in detail specific embodiments of electric vehicle charging station ordered charging methods based on a double-layer optimization algorithm with reference to the drawings.

FIG. 1 is a sequential charging control framework for an electric vehicle charging station, which mainly comprises a power distribution network Management layer, a charging station operation layer and an electric vehicle user layer, wherein a perfect communication network is arranged among the three layers so as to perform necessary information interaction, a control core of the power distribution network Management layer is a Centralized Management System (CMS), the CMS comprises a scheduling program and a power flow calculation program, the scheduling program controls charging loads of all charging stations in any time period, the power flow calculation program checks whether voltage and current constraints are out of bounds or not, the charging station operation layer is provided with an intelligent man-machine interaction terminal, provides charging mode selection for inbound users and calculates service time (sum of charging time and queuing waiting time) of corresponding modes, transmits vehicle information to be charged to the CMS and receives scheduling, and the electric vehicle is provided with a vehicle-mounted intelligent terminal, so that the information of nearby charging stations can be checked and the station arrival time can be estimated through vehicle networking.

The charging station provides a charging mode in three as a premise for a user, and the charging mode is as follows:

(1) fast dc charging mode. The battery of the electric automobile is directly charged in a large current mode through the off-board charger. The time for fully charging the deeply discharged battery in the mode is about 2 hours, and the theoretical maximum time can be reduced to 0.5 hour;

(2) compared with the quick direct current charging, the slow charging power is lower, namely about 6.6 kW. charging current is only 16A, meanwhile, the charging time is prolonged, namely is 6-8 h;

(3) battery replacement mode. When the battery energy of the electric automobile is lower, the battery pack with low electric quantity is replaced by the battery pack with full electric quantity, so that the aim of rapidly supplementing the electric energy of the electric automobile is fulfilled. The battery replacement operation can be completed within 3min, and theoretically, the driving range of the electric automobile can be infinitely extended.

The following description is made for a specific control method:

an upper and lower layer scheduling model is established based on the control framework shown in fig. 1, an upper layer power distribution network management layer formulates a scheduling plan of each charging station, and a lower layer charging station operation layer determines a specific charging scheme of each electric vehicle according to the upper layer scheduling plan.

(1) Upper optimization model

The upper layer model needs to consider that the fluctuation of the charging load is unfavorable for the operation of the power grid, the comprehensive minimum of the deviation of the dispatching plan is determined by the load variance of the power distribution network and the upper and lower layer models as an optimization target, and the objective function is as follows:

t is total time period number of scheduling middle periods, P_t ^aThe conventional residential electricity load is a time period t; p_j,tThe charging load of a charging station J is a time period t, and J is the number of the charging stations; p^avIs a system assemblyα is a weighting coefficient which represents the punishment to the upper and lower layers of the dispatch plan deviation;

and charging corresponding loads of the charging station j in the time period t obtained by optimization for the lower layer.

Node voltage amplitude constraints

U_m ^min≤U_m≤U_m ^max(2)

In the formula of U_m ^maxAnd U_m ^minRespectively, the upper and lower voltage limits of node m.

Line transmission power constraint

|P_l,t|≤P_l ^max(3)

In the formula P_l ^maxThe maximum operating transmission power for line l.

(2) Lower optimization model

The lower layer optimization is based on the premise that the deviation between the scheduling plan determined by the upper layer and the lower layer response charging load is minimum, and the objective function is as follows:

in the formula

Responding to the charging plan for the lower layer; i is the number of electric vehicles in the charging station j; x is the number of_i,k,tIs a decision variable, where a value equal to 1 indicates that the electric vehicle i is in a charging state in mode k for a period of time t, and a value equal to 0 indicates no charging. Wherein, k is 1 for fast charging mode, k is 2 for slow charging mode, and k is 3 for changing battery mode.

The lower layer model calculates a response charging plan through formula 5 and feeds the response charging plan back to the upper layer, and the upper layer carries out optimization calculation again until the convergence criterion is met:

in the formula, epsilon is convergence precision, and the invention takes 10^-3。

Maximum driving mileage constraint of electric automobile

In the formula, E_i ^BI, the rated capacity of a battery of the electric automobile; SOC_i,tThe SOC of the electric vehicle i is a time period t;

the power consumption of the electric automobile is per hundred miles.

Electric vehicle SOC constraint

SOC_i ^min≤SOC_i,t≤SOC_i ^max(8)

In the formula, SOC_i ^max、SOC_i ^minThe SOC upper and lower limits of the electric automobile are respectively 1 and 0.1.

FIG. 2 is a flow chart of an electric vehicle ordered charging solving based on a double-layer optimization model, the double-layer optimization model is solved by adopting a layered iteration method, and an optimal charging plan of a charging station and an optimal charging scheme of an electric vehicle can be obtained by repeatedly iterating an upper layer model and a lower layer model until constraints are finally met. Firstly, the upper layer adopts a genetic algorithm to schedule a plan P for each charging station_j,tEncoding to form an initial population, performing fitness function F, and performing algorithm iteration to finally obtain P_j,tAnd (5) optimal solution. The optimal solution is then communicated to the charging stations of the lower tier.

The lower layer considers a charging path selection mechanism of the electric automobile and adopts an ant colony algorithm to solve. Firstly, inputting parameters, position information, charging station information and the like of the electric vehicle, and calculating a minimum dispatching deviation function f of a lower layer model₁And based on the minimum deviation and the constraint of the formula 7-8, calculating the objective function f by adopting an ant colony algorithm. Form the optimal charging scheme of the electric automobile and will

And returning to the upper layer to continue optimization. And solving the terminating condition that the upper and lower layer scheduling deviation meets the convergence criterion.

Finally, it should be noted that the above embodiments are only used for illustrating the technical solutions of the present invention and not for limiting the same. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.