Charging station electric vehicle peak clipping response capability evaluation method considering vehicle pile matching relationship

Abstract

The invention discloses a charging station electric vehicle peak clipping response capability evaluation method considering a vehicle-pile matching relationship. The method comprises the steps of constructing an electric vehicle load model and a future deployable charging pile model; designing a matching constraint rule of the electric vehicle and the charging pile; designing a multi-vehicle charging period non-overlapping constraint condition and carrying out linear processing; designing a power constraint condition of the charging pile and a charging time constraint condition of the electric vehicle; fully guaranteeing the charging demand of a user is taken as a primary target, an electric vehicle optimization scheduling model is established and solved in combination with reduction of the charging cost, and an optimal scheduling result is obtained; and establishing a charging station peak clipping response capability evaluation model, and solving the charging station peak clipping response capability evaluation model to obtain the relationship between the response quantity and the response cost. According to the method, the electric vehicle load translation constraints under different peak clipping response capacities can be considered, and the electric vehicle peak clipping response capability evaluation model is established, so that the model better fits an actual control scene.

Description

Technical Field

[0001] The present invention relates to an electric vehicle optimization scheduling technology, and in particular to a method for evaluating the peak-shaving response capability of electric vehicles at a charging station taking into account the vehicle-pile matching relationship. Background Art

[0002] With the large-scale access of electric vehicles and charging piles to the power grid, traditional distribution networks face challenges such as uneven temporal and spatial distribution of charging loads and difficulty in dynamically matching regulatory resources. It is urgent to establish a refined evaluation system to tap the flexible adjustment potential of electric vehicle clusters.

[0003] Currently, research on electric vehicles, both domestically and internationally, still needs improvement in terms of modeling sophistication and alignment with practical applications. First, there's the issue of model simplification. Most studies model electric vehicles as energy storage systems with continuously adjustable power. The charging power is often directly taken as the vehicle's power, and only the overall capacity constraints of the charging station are considered. However, in reality, the charging power and method of electric vehicles must be matched to the charging piles, and they currently lack reverse discharge capabilities. Consequently, their control characteristics are not entirely equivalent to those of energy storage. Charging piles themselves have limitations, such as power and charging method. AC piles are typically 7kW, while DC piles range from 20 to 160kW. The charging power supported by commercially available electric vehicles varies, and some models do not support DC fast charging. Direct scheduling without considering the matching of vehicles and charging piles can result in charging failures, rendering the scheduling plan ineffective. Second, the optimal scheduling timeframe is relatively crude. Furthermore, most studies divide the day into several equal time periods for discrete scheduling, which underutilizes charging piles and increases unnecessary waiting time for users. Therefore, more realistic and refined modeling is needed for research on optimal electric vehicle scheduling.

[0004] Due to their high charging power and flexibility, electric vehicles have the potential to respond to demand in the power grid, supporting services such as frequency regulation and peak-load shifting. Existing demand response models have insufficiently evaluated the demand response capabilities of electric vehicles at charging stations. These models typically consider only the available capacity or use a fixed elasticity coefficient to reflect the response cost. These models lack a thorough analysis of the relationship between the amount of regulation and the cost, and overlook the limitations imposed by the number of charging piles available at charging stations. Consequently, these demand response models are not well-suited to practical applications. Summary of the Invention

[0005] Purpose of the invention: The purpose of the present invention is to provide a method for evaluating the peak-shaving response capability of electric vehicles at charging stations that takes into account the matching relationship between vehicles and charging piles. It can consider the load translation constraints of electric vehicles under different peak-shaving response capacities, establish an electric vehicle peak-shaving response capability evaluation model, and make the model more in line with actual control scenarios.

[0006] Technical Solution: The present invention provides a method for evaluating the peak-shaving response capability of electric vehicles at charging stations, taking into account the vehicle-pile matching relationship, including:

[0007] Use time-domain continuous variables to finely characterize the charging time period of electric vehicles and build an electric vehicle load model; build a future adjustable charging pile model based on the restrictions on the charging time period of charging piles;

[0008] Considering the number and model of available charging piles, the vehicle-pile correspondence is represented by a matrix to design matching constraint rules for electric vehicles and charging piles. Non-overlapping constraints for multiple vehicle charging periods are designed, and linearization technology is used to handle non-overlapping constraints for multiple vehicle charging periods. Power constraints for charging piles and charging time constraints for electric vehicles are designed.

[0009] Based on the electric vehicle load model and the future deployable charging pile model, the matching constraints between electric vehicles and charging piles and the non-overlapping charging period constraints of multiple vehicles are considered. With the primary goal of fully ensuring the charging needs of users, combined with reducing charging costs, an optimal scheduling model for electric vehicles is established and solved to obtain the optimal scheduling result.

[0010] Based on the optimal scheduling results, considering the electric vehicle load translation constraints under different peak shaving response capacities, a charging station peak shaving response capacity evaluation model is established. The relationship between the response amount and the response cost is obtained by solving the charging station peak shaving response capacity evaluation model.

[0011] Furthermore, the time domain continuous variables are used to finely characterize the charging time period of electric vehicles and construct an electric vehicle load model, including:

[0012] For each vehicle waiting to be serviced in the charging station, a multi-dimensional state vector is used for complete representation:

[0013] γ i =(s i ,E i ,P max,i ,η i ,SOC S,i ,SOC E,i ,SOC i,max ,SOC i,min ,T S,i ,T E,i )

[0014] Where i is the electric vehicle number; γ i is the multidimensional state vector of electric vehicle i; s i E is the binary charging status identifier of electric vehicle i; i is the total capacity of the power battery pack in electric vehicle i; P max,i is the maximum charging power allowed for electric vehicle i; η i The charging efficiency of electric vehicle i; SOC S,iis the current charge level of electric vehicle i; SOC E,i The charging termination target value of electric vehicle i; SOC i,max and SOC i,min are the upper and lower limits of battery state of charge respectively; T S,i and T E,i are the start and end times of charging of electric vehicle i, respectively, and support the use of continuous time expression;

[0015] Based on the principle of matching a vehicle with a single charging pile, the effective charging period of the vehicle can be expressed as:

[0016] T a,i =[T S,i ,T E,i ]∩[T cS,j ,T cE,j ]

[0017] Among them, T a,i is the effective charging interval after electric vehicle i is matched with the charging pile; j is the charging pile number matching electric vehicle i; T cS,j is the adjustable start time for matching charging pile j; T cE,j To match the adjustable end time of charging pile j, T cS,j and T cE,j Supports continuous time representation.

[0018] Furthermore, a model of future adjustable charging piles is constructed based on the restrictions on the charging time period of the charging piles, including:

[0019] At any scheduling time t1, if vehicle i is already in the charging state, then s i =1, indicating that the electric vehicle has been dispatched and will be classified as an undispatched load at and after the dispatch time t1. Combined with the historical dispatch plans before the dispatch time t1, the following set of future dispatchable charging piles can be obtained, which is used as the future dispatchable charging pile model:

[0020]

[0021] Where Γ1 is the set of available charging piles at the current scheduling time node t1; K is the number of newly added available charging pile events; Γ k is the scheduling time node t k Newly added adjustable charging pile set to meet k=2,...,K;n k is the scheduling time node t k Increase the number of charging piles that can be deployed; kj is the scheduling time node t k The jth newly added adjustable charging pile, k=2,...,K, j=1,...,nk ;

[0022] When there is a vehicle at the dispatch time node t k When the charging task is completed, the new deployable charging pile will be released; through the above formula, the set of all deployable charging piles after the current dispatch time node t1 can be obtained:

[0023]

[0024] Where Γ is the set of all deployable charging piles after the dispatch time node t1; p j is the deployable charging pile numbered j, j = 1,...,N c ; N c is the total number of all deployable charging piles after the dispatch time node t1.

[0025] For charging pile j, its different deployable time periods will be regarded as different elements in the resource set Γ, and its deployable time period is as follows:

[0026] T c,j = [T cS,j , T cE,j ], j = 1, 2,..., N c

[0027] Where T c,j is the deployable time period of charging pile j; T cS,j is the deployable start time of matching charging pile j; T cE,j is the deployable end time of matching charging pile j.

[0028] Further, considering the number and type of deployable charging piles, the vehicle-pile correspondence relationship is represented by a matrix, so as to design the matching constraint rules of electric vehicles and charging piles, including:

[0029] According to the principle of one vehicle and one pile, an N v × N c mapping matrix is constructed to represent the vehicle-charging pile matching relationship:

[0030]

[0031] Where N v is the number of electric vehicles to be dispatched; x ij is the i-th element in the i-th row of the matrix, i = 1,...,N v , j = 1,...,N c , x ij takes 1 when electric vehicle i and charging pile i establish a matching relationship, otherwise takes 0.

[0032] When a charging station is operating at high load, there is a situation where a vehicle cannot obtain matching charging resources, resulting in a failure to charge. At the same time, following the principle of one vehicle per charging pile, the following formula can be used to ensure that a single electric vehicle can be assigned to at most one charging pile:

[0033]

[0034] Furthermore, a non-overlapping constraint condition for charging periods of multiple vehicles is designed, and a linearization technique is used to process the non-overlapping constraint condition for charging periods of multiple vehicles, including:

[0035] In different time periods, the same charging pile can provide services for multiple vehicles, but must meet the mutual exclusion constraint in the time dimension; when charging pile j is charging electric vehicle i and electric vehicle l, its service time period must meet the following requirements:

[0036] (t S,i -t E,l )(t E,i -t S,l )≥0

[0037] Among them, t S,i is the time when electric vehicle i starts charging; t E,i The end time of charging of electric vehicle i; t E,l The charging end time of electric vehicle l; t S,l The time when charging of electric vehicle l starts;

[0038] The nonlinear terms are converted into mixed integer linear forms by the big M method:

[0039]

[0040] Among them, y ilj is an auxiliary 0-1 variable; M is a sufficiently large positive constant;

[0041] When two electric vehicles share a charging pile, the charging time period of electric vehicle i (t S,i ,t E,i ) and the charging time period of electric vehicle l (t S,l ,t E,l ) These two non-intersecting constraints should take effect, from which we can deduce:

[0042]

[0043] Furthermore, the power constraints of the charging pile and the charging time constraints of the electric vehicle are designed, including:

[0044] The actual charging power of an electric vehicle is determined by the rated output capacity of the matching charging pile and should not exceed the vehicle's own power tolerance limit:

[0045]

[0046] Where: P c,j is the maximum output power of charging pile j; N c is the total number of all available charging piles after the scheduling time node t1; max,i is the maximum charging power allowed for electric vehicle i; N v is the number of electric vehicles to be dispatched;

[0047] The charging period for electric vehicles must be strictly limited to the service time of the equipment. That is, the charging start time must not be earlier than the charging pile activation time, and the charging end time must not be later than the charging pile deactivation time:

[0048]

[0049] Furthermore, based on the electric vehicle load model and the future deployable charging pile model, the matching constraints between electric vehicles and charging piles, the non-overlapping constraints of multiple vehicle charging periods, the power constraints of charging piles, and the charging time constraints of electric vehicles are considered. With the primary goal of fully ensuring the charging needs of users, combined with reducing charging costs, an electric vehicle optimization scheduling model is established and solved to obtain the optimal scheduling results, including:

[0050] Electric vehicle scheduling adopts a multi-objective optimization framework, giving priority to ensuring the satisfaction of charging demand, considering the reduction of total cost, and transforming the multi-objective model into a single-objective model through the weighted coefficient method. The objective function of the electric vehicle optimization scheduling model is:

[0051]

[0052] Among them, f is the comprehensive optimization objective function; w1 and w2 are weight coefficients, satisfying w1>>w2; r is the slope function; E i is the total capacity of the power battery pack in electric vehicle i; SOC S,i is the current charge level of electric vehicle i; SOC E,i is the charging termination target value of electric vehicle i; t S,i is the time when electric vehicle i starts charging; t E,i is the end charging time of electric vehicle i; t is time; c t is the electricity price at time t; P c,j is the maximum output power of charging pile j.

[0053] Furthermore, based on the electric vehicle load model and the future deployable charging pile model, the matching constraints between electric vehicles and charging piles, the non-overlapping constraints of multiple vehicle charging periods, the power constraints of charging piles, and the charging time constraints of electric vehicles are considered. With the primary goal of fully ensuring the charging needs of users, combined with reducing charging costs, an optimal scheduling model for electric vehicles is established and solved, and the optimal scheduling results are obtained, including:

[0054] When solving the electric vehicle optimization scheduling model, the matching constraints between electric vehicles and charging piles must be followed. In addition, vehicle charging operations must be completed within the user-specified allowed charging time period, that is, the actual charging period must be completely included in the allowed time period:

[0055] T S,i ≤t S,i ≤t E,i ≤T E,i ,i=1,2,…,N v

[0056] The total amount of charging should not exceed the required value preset by the user:

[0057] P v,i (t E,i -t S,i )≤E i (SOC E,i -SOC S,i ),i=1,2,…,N v

[0058] Among them, E i is the total capacity of the power battery pack in electric vehicle i; SOC S,i is the current charge level of electric vehicle i; SOC E,i is the charging termination target value of electric vehicle i; P v,i is the actual charging power of electric vehicle i, which is determined by the power of the corresponding charging pile that matches the electric vehicle. Its expression is as follows:

[0059]

[0060] The end time of electric vehicle charging must be later than the start time:

[0061] 0≤t S,i ≤t E,i ,i=1,2,…,N v .

[0062] Among them, t S,i is the time when electric vehicle i starts charging; t E,i The moment of final charging for the electric car i.

[0063] Furthermore, based on the optimal scheduling results, considering the electric vehicle load shift constraints under different peak shaving response capacities, a charging station peak shaving response capacity evaluation model is established. The relationship between the response amount and the response cost is obtained by solving the charging station peak shaving response capacity evaluation model, including:

[0064] According to the charging start time and end time of each electric vehicle in the optimal scheduling result, the charging load curve of the electric vehicles at the charging station is obtained as shown below:

[0065]

[0066] Among them, P t is the load of the charging station at time t; S,i is the time when electric vehicle i starts charging; t E,i The charging end time of electric vehicle i; P v,i is the actual charging power of electric vehicle i;

[0067] Considering the electric vehicle load shift constraints under different peak shaving response capacities: the load P of the charging station at time t is calculated as follows: t :

[0068] For all t S,i and t E,i Sort from small to large and store them in an array. Use variable m to represent the sequence number of the elements in the array, and record the elements in the array as t. l,m ; Establish a set Γ to store the electric vehicle loads that need to be accumulated;

[0069] Let m = 1, then t l,m Corresponding to the variable t at the start time of charging of an electric vehicle S,i , the electric vehicle load P v,i Add to the set Γ;

[0070] Let m=m+1, calculate the charging station in the period [t l,m-1 ,t l,m ) is:

[0071]

[0072] If t l,m Corresponding to a variable t representing the starting time of electric vehicle charging S,i , then the load P v,i Add to the set Γ; otherwise, t l,m is a variable t corresponding to the time when the electric vehicle charging ends E,i , then remove the element P from the set Γ v,i ;

[0073] If tl,m If it is the last element in the array, the charging load curve calculation is completed, otherwise the charging station is recalculated in the period [t l,m-1 ,t l,m ) load.

[0074] Furthermore, based on the optimal scheduling results and considering the electric vehicle load shift constraints under different peak shaving response capacities, a charging station peak shaving response capacity evaluation model is established. The relationship between the response amount and the response cost is obtained by solving the charging station peak shaving response capacity evaluation model, including:

[0075] According to the matching relationship matrix X between electric vehicles and charging piles, the available time period T for each charging pile after scheduling can be obtained: c,j :

[0076]

[0077] Among them, T cS,j is the adjustable start time for matching charging pile j; T cE,j To match the end time of the charging pile j; t S,i is the time when electric vehicle i starts charging; t E,i The moment of charging the electric car i ends;

[0078] Changing the charging period of electric vehicles as a way for electric vehicles to participate in peak shaving response, sorting the charging loads of each electric vehicle in the peak shaving response period, selecting electric vehicles to participate in peak shaving response in the order of gradually increasing peak shaving response amount, and rescheduling the electric vehicles;

[0079] The optimal scheduling considering peak shaving response not only includes the constraints of the electric vehicle optimal scheduling model, but also includes the constraint that the selected electric vehicles are not charged during the peak shaving response period, which can be expressed as follows:

[0080] (t S,i -t Ed )(t E,i -t Sd )≥0

[0081] Among them, t Sd and t Ed are the start and end times of the peak clipping response respectively;

[0082] The above formula is converted into mixed integer linear form by the big M method as follows:

[0083]

[0084] Among them, y i is an auxiliary 0-1 variable;

[0085] The cost of peak load response at a charging station consists of two parts: the change in charging electricity charges and the compensation for the reduction in user charging. The following formula is used to measure the cost of demand response at a charging station:

[0086]

[0087] Where C is the peak shaving response cost of the charging station; c e,t is the time-of-use electricity price; c m,t is the charging price of the charging pile during the peak electricity price period; t o,S,i and t o,E,i are the charging start and end time of electric vehicle i before peak shaving response; P v,i is the actual charging power of electric vehicle i; N v is the number of electric vehicles to be dispatched;

[0088] The peak shaving response amount is gradually increased, the above optimization problems are solved in sequence, and the corresponding charging station response costs are calculated, thereby establishing a relationship between the peak shaving response amount and the response cost of the charging station.

[0089] Beneficial effects: Compared with the existing technology, the significant technical effects of the present invention are as follows: (1) In the evaluation of the peak-shaving response capability of electric vehicles at charging stations, the matching situation between electric vehicles and charging piles is taken into account, and the obtained results are more in line with the actual situation and have the feasibility of application implementation; (2) The introduction of time-domain continuous variables and dynamic charging pile set models breaks through the limitations of traditional discrete time division, accurately characterizes the timing characteristics of charging pile resource release, and solves the problem of underestimated regulation capability caused by insufficient resource utilization efficiency; (3) In the analysis of the peak-shaving response capability of charging stations, the restrictions of adjustable charging piles on regulation are taken into account, and the relationship between response amount and response cost is obtained, which can evaluate the peak-shaving response capability of electric vehicles at charging stations in a more comprehensive and practical way; (4) The present invention can deeply release the two-way interactive potential of electric vehicles and smart grids, build a flexible regulation system for vehicle-grid collaboration, and provide dynamic response support for the low-carbon evolution of energy systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] Figure 1 It is a schematic diagram of the process of the present invention;

[0091] Figure 2 This is a curve diagram of demand response cost and the number of vehicles participating in the response as the peak shaving response amount changes. DETAILED DESCRIPTION

[0092] The technical solution of the present invention is described in detail below in conjunction with specific implementation methods and the accompanying drawings.

[0093] like Figure 1 As shown, a method for evaluating the peak-shaving response capability of electric vehicles at a charging station taking into account the vehicle-pile matching relationship of the present invention includes the following steps:

[0094] S1. Use time-domain continuous variables to finely characterize the charging time period of electric vehicles and build an electric vehicle load model; build a future adjustable charging pile model based on the restrictions on the charging time period of charging piles.

[0095] The specific implementation process of step S1 is as follows:

[0096] S1.1. Use time-domain continuous variables to finely characterize the charging time period of electric vehicles and construct an electric vehicle load model, as follows:

[0097] The load characteristic modeling of electric vehicles requires comprehensive consideration of inherent parameters such as energy storage capacity, nominal charging power, and conversion efficiency, while also incorporating dynamic variables such as available charging time and real-time charging status. For each vehicle waiting to be serviced at the charging station, a multi-dimensional state vector is used for complete representation:

[0098] γ i =(s i ,E i ,P max,i ,η i ,SOC S,i ,SOC E,i ,SOC i,max ,SOC i,min ,T S,i ,T E,i )

[0099] Where i is the electric vehicle number; γ i is the multidimensional state vector of electric vehicle i; s i is the binary charging status identifier of electric vehicle i (1 when charging, 0 otherwise); E i is the total capacity of the power battery pack in electric vehicle i; P max,i is the maximum charging power allowed for electric vehicle i; η i is the charging efficiency of electric vehicle i; SOC S,i is the current charge level of electric vehicle i; SOC E,i The charging termination target value of electric vehicle i; SOC i,max and SOC i,min are the upper and lower limits of battery state of charge respectively; T S,i and T E,i They are the start and end times of charging of electric vehicle i, and support the use of continuous time expression.

[0100] The above formula can fully describe the state of the vehicle itself. It is necessary to further establish a matching mechanism between the vehicle and the charging pile to generate the actual load. To avoid repeated scheduling and ineffective movement of vehicles within the station, the principle of matching the vehicle with a single charging pile is adopted. In this case, the effective charging period of the vehicle can be expressed as:

[0101] T a,i =[T S,i ,T E,i ]∩[T cS,j ,T cE,j ]

[0102] Among them, T a,i is the effective charging interval after electric vehicle i is matched with the charging pile; j is the charging pile number matching electric vehicle i; T cS,j is the adjustable start time for matching charging pile j; T cE,j To match the adjustable end time of charging pile j, T cS,j and T cE,j Supports continuous time representation.

[0103] S1.2. Construct a model of future deployable charging piles based on the charging time limit of the charging piles. The details are as follows:

[0104] Since the effective charging period of electric vehicles is limited by the service time period of charging piles, it is necessary to construct a future deployable charging pile model based on the restrictions on the charging time period of charging piles.

[0105] To avoid repeated dispatching and invalid movement of vehicles in the station, at any dispatching time t1, if vehicle i is already in the charging state, then s i =1, indicating that the electric vehicle has been dispatched and will be classified as an undispatched load at and after the dispatch time t1. Combined with the historical dispatch plan before the dispatch time t1, the following set of future dispatchable charging piles can be obtained, which is used as the future dispatchable charging pile model:

[0106]

[0107] Where Γ1 is the set of available charging piles at the current scheduling time node t1; K is the number of newly added available charging pile events; Γ k is the scheduling time node t k Newly added adjustable charging pile set to meet k=2,...,K;n k is the scheduling time node t k Increase the number of charging piles that can be deployed; kj is the scheduling time node t k The jth newly added adjustable charging pile, k=2,...,K, j=1,...,n k.

[0108] When a vehicle is dispatched at time node t k When the charging task is completed, the new available charging pile will be released. Through the above formula, we can get the set of all available charging piles after the current scheduling time node t1:

[0109]

[0110] Where Γ is the set of all available charging piles after the scheduling time node t1; p j is the adjustable charging pile numbered j, j = 1,...,N c ; N c It is the total number of all available charging piles after the scheduling time node t1.

[0111] In scheduling modeling, the temporal availability of charging pile resources is discretized into non-overlapping time periods. For charging pile j, its different available time periods will be treated as different elements in the resource set Γ. Its available time periods are as follows:

[0112] T c,j =[T cS,j ,T cE,j ],j=1,2,…,N c

[0113] Among them, T c,j is the available time period for charging pile j.

[0114] S2. Considering the number and models of available charging piles, the correspondence between vehicles and piles is represented by a matrix, thereby designing matching constraint rules for electric vehicles and charging piles; designing non-overlapping constraints for charging periods of multiple vehicles, and using linearization technology to handle non-overlapping constraints for charging periods of multiple vehicles; designing power constraints for charging piles and charging time constraints for electric vehicles.

[0115] The specific implementation process of step S2 is as follows:

[0116] S2.1. Considering the number and types of available charging piles, the vehicle-pile correspondence is represented by a matrix, thereby designing the matching constraint rules between electric vehicles and charging piles. The specific rules are as follows:

[0117] Given that each electric vehicle can only use one charging pile for charging at the same time, and the charging process of an electric vehicle using a charging pile must comply with corresponding operating constraints, it is necessary to characterize the vehicle-charging pile matching relationship.

[0118] The service capacity of the charging station is restricted by the number and type of charging piles in the station. During high-load periods, intelligent scheduling is required to optimize the configuration of vehicle demand and facility resources to meet charging demand while reducing charging costs. During low-load periods, automatic matching of vehicles and charging piles can be performed. Following the principle of one vehicle per pile (i.e., one electric vehicle can only occupy one charging pile), N v ×N c The mapping matrix represents the vehicle-charging pile matching relationship:

[0119]

[0120] Among them, N v is the number of electric vehicles to be dispatched; x ij is the i-th element in the i-th row of the matrix, i=1,...,N v ,j=1,...,N c , x ij is a binary mapping variable, x ij When it is 1, it means that electric vehicle i and charging pile i have established a matching relationship, otherwise it is 0.

[0121] When a charging station is operating at high load, it is possible that a vehicle cannot obtain matching charging resources, resulting in a failure to charge. At the same time, following the principle of one vehicle per charging station, the following formula can be used to ensure that a single electric vehicle can only be assigned to one charging station at most:

[0122]

[0123] S2.2. Design the non-overlapping constraint conditions for multiple vehicle charging periods and use linearization technology to handle the non-overlapping constraint conditions for multiple vehicle charging periods. The details are as follows:

[0124] Design the non-overlapping charging period constraint: The same charging pile can provide services for multiple vehicles in different time periods, but must meet the mutual exclusion constraint in the time dimension; when charging pile j is charging electric vehicle i and electric vehicle l, its service time period must meet the following conditions:

[0125] (t S,i -t E,l )(t E,i -t S,l )≥0

[0126] Among them, t S,i is the time when electric vehicle i starts charging; t E,i The charging end time of electric vehicle i; t E,l The charging end time of electric vehicle l; t S,l The time when electric vehicle l starts charging.

[0127] It is easy to verify that under the above constraints (tS,i ,t E,i ) and (t S,l ,t E,l ) The two intervals do not intersect.

[0128] Linearization technology is used to deal with the non-overlapping constraint of multiple vehicle charging periods: To improve the solution efficiency, the nonlinear terms are converted into mixed integer linear forms using the big M method:

[0129]

[0130] Among them, y ilj is an auxiliary 0-1 variable; M is a sufficiently large positive constant.

[0131] In the above formula, when y ilj =1, we get (t S,i -t E,l )≤0、(t E,i -t S,l )≤0, that is, the end time of charging of electric vehicle i is earlier than the start time of charging of electric vehicle l; when y ilj = 0, we get (t S,i -t E,l )≥0、(t E,i -t S,l )≥0, that is, the start charging time of electric vehicle i is later than the end charging time of electric vehicle l. This shows that the charging time of two electric vehicles using the same charging pile does not overlap. Therefore, the constraint shown in the above formula is equivalent to (t S,i -t E,l )(t E,i -t S,l )≥0.

[0132] When the variable x ij and x lj When both are 1, it means that the i-th and l-th electric vehicles are assigned to the same charging station j. In this case, the non-overlapping constraint in the time dimension needs to be activated, that is, the charging time period of electric vehicle i (t S,i ,t E,i ) and the charging time period of electric vehicle l (t S,l ,t E,l ) must meet the mutually exclusive condition. When two electric vehicles share a charging pile (i.e. x ij =x lj =1), the charging time period of electric vehicle i (t S,i ,t E,i ) and the charging time period of electric vehicle l (t S,l ,t E,l ) These two non-intersecting constraints should take effect, from which we can deduce:

[0133]

[0134] S2.3. Design the power constraints of the charging pile and the charging time constraints of the electric vehicle.

[0135] Charging pile power constraints: The actual charging power of an electric vehicle is determined by the rated output capacity of the matching charging pile and should not exceed the vehicle's own power tolerance limit:

[0136]

[0137] Among them, P c,j is the maximum output power of charging pile j; N c is the total number of all available charging piles after the scheduling time node t1; max,i is the maximum charging power allowed for electric vehicle i.

[0138] Electric vehicle charging time constraints: Electric vehicle charging periods must be strictly limited to the equipment service time, that is, the charging start time must not be earlier than the charging pile activation time, and the charging end time must not be later than the charging pile deactivation time:

[0139]

[0140] Among them, T cS,j is the adjustable start time for matching charging pile j; T cE,j To match the adjustable end time of charging pile j.

[0141] S3. Based on the electric vehicle load model and the future adjustable charging pile model, considering the matching constraints between electric vehicles and charging piles, the non-overlapping constraints of multiple vehicle charging periods, the power constraints of charging piles and the charging time constraints of electric vehicles, with the primary goal of fully ensuring the charging needs of users, combined with reducing charging costs, an electric vehicle optimization scheduling model is established and solved to obtain the optimal scheduling result.

[0142] The specific implementation process of step S3 is as follows:

[0143] S3.1. Establish an electric vehicle optimization scheduling model.

[0144] In actual operation, the charging power of electric vehicles needs to match the charging pile, and it is usually the rated value under existing technical and economic conditions. Therefore, the core of the electric vehicle optimization scheduling decision lies in optimizing the charging period, and this type of load can be regarded as a shiftable load. Under the rigid constraint of travel demand, compared with the fluctuation of electricity prices in different periods, users are more inclined to quickly obtain charging piles that match the vehicle parameters to shorten the waiting time. Therefore, electric vehicle scheduling adopts a multi-objective optimization framework, in which priority is given to ensuring the satisfaction of charging demand, and secondly considering the reduction of total cost. The multi-objective model is converted into a single-objective model through the weighted coefficient method. The objective function of the electric vehicle optimization scheduling model is:

[0145]

[0146] Where f is the comprehensive optimization objective function; w1 and w2 are weight coefficients, satisfying w1>>w2, that is, by increasing the cost of violating the first sub-goal to guide the optimization process towards the direction of satisfying the first sub-goal, to maximize the satisfaction of user charging needs; r is the ramp function, which is 0 when the actual charging amount exceeds the user's expected charging amount, avoiding sub-goal 1 from tending to negative infinity; E i is the total capacity of the power battery pack in electric vehicle i; SOC S,i is the current charge level of electric vehicle i; SOC E,i is the charging termination target value of electric vehicle i; t S,i is the time when electric vehicle i starts charging; t E,i is the end charging time of electric vehicle i; t is time; P c,j is the maximum output power of charging pile j; c t is the electricity price at time t, which is composed of the time-of-use electricity price and service fee, as shown in the following formula:

[0147] c t =c e,t +c s

[0148] Where: c e,t is the time-of-use electricity price; c s The price of charging service at the charging station.

[0149] According to the peak and valley electricity prices, the cost integral term in the objective function is converted into a piecewise expression:

[0150]

[0151] Among them, t hS , t hE are the start and end time of the peak electricity price respectively; c h 、c l are the peak and valley electricity prices respectively; r lis the limiting ramp function, and its mathematical expression is:

[0152]

[0153] Where z is a real variable.

[0154] S3.2. When solving the optimal scheduling model for electric vehicles, the matching constraints between electric vehicles and charging piles in step S2 must be followed. In addition, the vehicle charging operation must be completed within the user-specified allowed charging time period, that is, the actual charging period should be completely included in the allowed time period:

[0155] T S,i ≤t S,i ≤t E,i ≤T E,i ,i=1,2,…,N v

[0156] The total amount of charging should not exceed the required value preset by the user:

[0157] P v,i (t E,i -t S,i )≤E i (SOC E,i -SOC S,i ),i=1,2,…,N v

[0158] Among them, E i is the total capacity of the power battery pack in electric vehicle i; SOC S,i is the current charge level of electric vehicle i; SOC E,i is the charging termination target value of electric vehicle i; P v,i is the actual charging power of electric vehicle i, which is determined by the power of the corresponding charging pile that matches the electric vehicle. Its expression is as follows:

[0159]

[0160] The end time of electric vehicle charging must be later than the start time:

[0161] 0≤t S,i ≤t E,i ,i=1,2,…,N v

[0162] Among them, t S,i is the time when electric vehicle i starts charging; t E,i The moment of final charging for electric car i.

[0163] S4. Based on the optimal scheduling results, considering the electric vehicle load translation constraints under different peak-shaving response capacities, a charging station peak-shaving response capacity evaluation model is established, and the relationship between the response amount and the response cost is obtained by solving the charging station peak-shaving response capacity evaluation model.

[0164] S4.1. Based on the charging start and end times of each electric vehicle in the optimal scheduling results, the charging load curve of the electric vehicles at the charging station can be obtained as shown in the following formula:

[0165]

[0166] Among them, P t is the load of the charging station at time t; P v,i is the actual charging power of electric vehicle i.

[0167] Considering the load shift constraints of electric vehicles under different peak shaving response capacities: The above formula can be calculated in the actual program as follows:

[0168] ① For all t S,i and t E,i Sort from small to large and store them in an array. Use variable m to represent the sequence number of the elements in the array, and record the elements in the array as t. l,m A set Γ is established to store the electric vehicle loads that need to be accumulated.

[0169] ② Let m = 1, then t l,m Corresponding to the variable t at the start time of charging of an electric vehicle S,i , the electric vehicle load P v,i Add to the set Γ.

[0170] ③ Let m=m+1, calculate the charging station in the time period [t l,m-1 ,t l,m ) is:

[0171]

[0172] ④ If t l,m Corresponding to a variable t representing the starting time of electric vehicle charging S,i , then the load P v,i Add to the set Γ; otherwise, t l,m is a variable t corresponding to the time when the electric vehicle charging ends E,i , then remove the element P from the set Γ v,i .

[0173] ⑤If t l,m If it is the last element in the array, the load curve calculation is completed, otherwise return to step ③ to continue the calculation.

[0174] S4.2. Establish a charging station peak shaving response capability assessment model.

[0175] At present, the peak load shaving and load regulation demand response of the power system is usually implemented during the peak load period of the day. As an important response resource on the load side, the peak load shaving response capability of the charging station in a given period needs to be analyzed. According to the matching relationship matrix X between electric vehicles and charging piles, the available time period T of each charging pile after scheduling can be obtained. c,j :

[0176]

[0177] Among them, T cS,j is the adjustable start time for matching charging pile j; T cE,j To match the end time of the charging pile j; t S,i is the time when electric vehicle i starts charging; t E,i The moment of final charging for electric car i.

[0178] Considering that the way electric vehicles participate in peak shaving response is mainly to change the charging period of electric vehicles, peak shaving response can be carried out by regulating the electric vehicles charging during the peak shaving response period.

[0179] The charging loads of the electric vehicles in the peak shaving response period are sorted from small to large, and the electric vehicles participating in the peak shaving response are selected in order of increasing peak shaving response amount, and the electric vehicles are rescheduled. The optimal scheduling considering the peak shaving response includes not only the constraints of the electric vehicle optimal scheduling model in step S3.2, but also the constraint that the selected electric vehicles are not charged during the peak shaving response period, which can be expressed as follows:

[0180] (t S,i -t Ed )(t E,i -t Sd )≥0

[0181] Among them, t Sd and t Ed are the start and end times of the peak clipping response respectively;

[0182] Following the similar large-M method in step S2.1, the above equation is converted into a mixed integer linear form as follows:

[0183]

[0184] Where: y i Auxiliary 0-1 variable.

[0185] The decision of a charging station to participate in peak load response is mainly influenced by the response subsidy price and the response cost. The evaluation of its peak load response capability is a description of the relationship between the response amount and the response cost. The cost of peak load response of a charging station is composed of two parts: the change in charging electricity price and the compensation for the user's reduced charging amount. Drawing on the demand response penalty mechanism of the Singapore electricity market, the user's reduced charging amount is compensated by the charging price during the peak electricity price period, so the demand response cost of the charging station is measured using the following formula:

[0186]

[0187] Where C is the peak shaving response cost of the charging station; c e,t is the time-of-use electricity price; c m,t is the charging price of the charging pile during the peak electricity price period; t o,S,i and t o,E,i are the charging start and end times of electric vehicle i before peak shaving response.

[0188] Since the charging price of the charging pile during the peak electricity price period is c m,t It is always higher than the time-of-use electricity price. Therefore, under this cost mechanism, charging stations will still be willing to meet users' charging needs as much as possible when participating in peak shaving response.

[0189] By gradually increasing the peak-shaving response amount, solving the above optimization problems in turn, and calculating the corresponding charging station response costs, the relationship between the peak-shaving response amount and the response cost of the charging station is established. Among them, the limitations of the charging piles that can be deployed at the charging station on the demand response are taken into account, which can more comprehensively and realistically evaluate the charging station's electric vehicle demand response capability.

[0190] The effect of the method for evaluating the peak-shaving response capability of electric vehicles at a charging station taking into account the vehicle-pile matching relationship of the present invention will be verified below through a specific embodiment.

[0191] The embodiment is set as a large charging station, which includes 15 charging piles and 150 electric vehicles to be dispatched. Referring to the typical electric vehicle parameters available on the market, the number ratio of electric vehicles and the probability of charging period in the embodiment are set as shown in the following table. The initial SOC is a random number between 0.1 and 0.3, the expected SOC is a random number between 0.8 and 0.9, and the maximum allowed charging power is one of 7 / 20 / 60 / 90 / 120 / 180kW. The rated power of the charging pile is 7 / 60 / 90 / 120 / 180kW. The scheduling cycle is 24 hours. It is set from 08:00 to 08:00 the next day based on the owner's travel and charging load rules. The electricity price during the peak period (08:00-22:00) is 1.15 yuan / kW·h, and the electricity price during the valley period (22:00-08:00) is 0.35 yuan / kW·h. The charging service fee of the charging station is uniformly set to 0.8 yuan / kW·h.

[0192]

[0193] The method of the present invention is used to obtain the demand response cost of the charging station during the period of 17:00-18:00 and the number of vehicles participating in the response as the peak shaving response amount changes. Figure 2 As shown:

[0194] Depend on Figure 2 As can be seen, after a certain point, as the response volume increases, the number of EVs participating in peak shaving gradually decreases. This is because different EVs have varying amounts of available response. Vehicles with low response volumes are selected first, requiring a larger number of vehicles. Then, vehicles with high response volumes are gradually selected, requiring a smaller number of vehicles. The cost of peak shaving response is approximately proportional to the response volume, as charging fees and user compensation are primarily based on the amount of charging.

[0195] Under the parameter settings in this example, the charging station's maximum response capacity between 5:00 PM and 6:00 PM is 392.6 kW·h, corresponding to a peak-shaving response cost of 348.7 yuan. Based on this evaluation model, combined with market factors such as the real-time peak-shaving response subsidy price and the terms of the charging agreement with the user, charging stations can make decisions on whether to participate in peak-shaving response and the specific amount of response to participate, contributing to the station's economic operation.

[0196] Based on the above analysis results, it can be seen that the present invention takes into account the restrictions on regulation of adjustable charging piles in the analysis of the peak-shaving response capability of charging stations, and obtains the relationship between the response amount and the response cost, which can more comprehensively and realistically evaluate the peak-shaving response capability of electric vehicles in charging stations.

Claims

1. A method for evaluating the peak load shaving response capability of electric vehicles at charging stations considering the matching relationship between vehicles and charging piles, characterized in that: include: Use time-domain continuous variables to finely characterize the charging time period of electric vehicles and build an electric vehicle load model; build a future adjustable charging pile model based on the restrictions on the charging time period of charging piles; Considering the number and model of available charging piles, the vehicle-pile correspondence is represented by a matrix to design matching constraint rules for electric vehicles and charging piles. Non-overlapping constraints for multiple vehicle charging periods are designed, and linearization technology is used to handle non-overlapping constraints for multiple vehicle charging periods. Power constraints for charging piles and charging time constraints for electric vehicles are designed. Based on the electric vehicle load model and the future deployable charging pile model, the matching constraints between electric vehicles and charging piles, the non-overlapping charging period constraints of multiple vehicles, the power constraints of charging piles, and the charging time constraints of electric vehicles are considered. With the primary goal of fully ensuring the charging needs of users, combined with reducing charging costs, an electric vehicle optimization scheduling model is established and solved to obtain the optimal scheduling result. Based on the optimal scheduling results, considering the electric vehicle load translation constraints under different peak shaving response capacities, a charging station peak shaving response capacity evaluation model is established. The relationship between the response amount and the response cost is obtained by solving the charging station peak shaving response capacity evaluation model.

2. The method for evaluating the peak load shaving response capability of electric vehicles at charging stations considering the vehicle-pile matching relationship according to claim 1 is characterized in that: The time domain continuous variables are used to finely characterize the charging time period of electric vehicles and construct an electric vehicle load model, including: For each vehicle waiting to be serviced in the charging station, a multi-dimensional state vector is used for complete representation: γ i =(s i ,E i ,P max,i ,η i ,SOC S,i ,SOC E,i ,SOC i,max ,SOC i,min ,T S,i ,T E,i ) Where i is the electric vehicle number; γ i is the multidimensional state vector of electric vehicle i; s i E is the binary charging status identifier of electric vehicle i; i is the total capacity of the power battery pack in electric vehicle i; P max,i is the maximum charging power allowed for electric vehicle i; η i is the charging efficiency of electric vehicle i; SOC S,i is the current charge level of electric vehicle i; SOC E,i The charging termination target value of electric vehicle i; SOC i,max and SOC i,min are the upper and lower limits of battery state of charge respectively; T S,i and T E,i are the start and end times of charging of electric vehicle i, respectively, and support the use of continuous time expression; Based on the principle of matching a vehicle with a single charging pile, the effective charging period of the vehicle can be expressed as: T a,i =[T S,i ,T E,i ]∩[T cS,j ,T cE,j ] Among them, T a,i is the effective charging interval after electric vehicle i is matched with the charging pile; j is the charging pile number matching electric vehicle i; T cS,j is the adjustable start time for matching charging pile j; T cE,j To match the available end time of charging pile j, T cS,j and T cE,j Supports continuous time representation.

3. The method for evaluating the peak load shaving response capability of electric vehicles at charging stations considering the vehicle-pile matching relationship according to claim 1 is characterized in that: Build a model of future adjustable charging piles based on the charging time period restrictions of the charging piles, including: At any scheduling time t1, if vehicle i is already in the charging state, then s i =1, indicating that the electric vehicle has been dispatched and will be classified as an undispatched load at and after the dispatch time t1. Combined with the historical dispatch plans before the dispatch time t1, the following set of future dispatchable charging piles can be obtained, which is used as the future dispatchable charging pile model: Where Γ1 is the set of available charging piles at the current scheduling time node t1; K is the number of newly added available charging pile events; Γ k is the scheduling time node t k Newly added adjustable charging pile set to meet k=2,...,K;n k is the scheduling time node t k Increase the number of charging piles that can be deployed; kj is the scheduling time node t k The jth newly added adjustable charging pile, k=2,...,K, j=1,...,n k ; When a vehicle is dispatched at time node t k When the charging task is completed, the new deployable charging pile will be released. Through the above formula, the set of all deployable charging piles after the current scheduling time node t1 can be obtained: Where Γ is the set of all available charging piles after the scheduling time node t1; p j is the adjustable charging pile numbered j, j = 1,...,N c ; N c is the total number of all configurable charging piles after the scheduling time node t1; for charging pile j, its different configurable time periods will be regarded as different elements in the resource set Γ, and its configurable time periods are as follows: T c,j =[T cS,j ,T cE,j ],j=1,2,…,N c Among them, T c,j is the available time period for charging pile j; T cS,j is the adjustable start time for matching charging pile j; T cE,j To match the adjustable end time of charging pile j.

4. The method for evaluating the peak load shaving response capability of electric vehicles at charging stations considering the vehicle-pile matching relationship according to claim 1 is characterized in that: Considering the number and models of available charging piles, the vehicle-pile correspondence is represented by a matrix, thereby designing matching constraint rules for electric vehicles and charging piles, including: Follow the principle of single bike and single pile, build N v ×N c The mapping matrix represents the vehicle-charging pile matching relationship: Among them, N v is the number of electric vehicles to be dispatched; x ij is the i-th element in the i-th row of the matrix, i=1,...,N v ,j=1,...,N c , x ij When it is 1, it means that the electric vehicle i and the charging pile i have established a matching relationship, otherwise it is 0; When a charging station is operating at high load, there is a situation where a vehicle cannot obtain matching charging resources, resulting in a failure to charge. At the same time, following the principle of one vehicle per charging pile, the following formula can be used to ensure that a single electric vehicle can be assigned to at most one charging pile:

5. The method for evaluating the peak load shaving response capability of electric vehicles at charging stations considering the vehicle-pile matching relationship according to claim 4 is characterized in that: Design the non-overlapping constraints for multiple vehicle charging periods and use linearization technology to handle the non-overlapping constraints for multiple vehicle charging periods, including: In different time periods, the same charging pile can provide services for multiple vehicles, but must meet the mutual exclusion constraint in the time dimension; when charging pile j is charging electric vehicle i and electric vehicle l, its service time period must meet the following requirements: (t S,i -t E,l )(t E,i -t S,l )≥0 Among them, t S,i is the time when electric vehicle i starts charging; t E,i The end time of charging of electric vehicle i; t E,l The charging end time of electric vehicle l; t S,l is the time when electric vehicle l starts charging; under the constraints of the above formula (t S,i ,t E,i ) and (t S,l ,t E,l ) The two intervals do not intersect; The nonlinear terms are converted into mixed integer linear forms by the big M method: Among them, y ilj is an auxiliary 0-1 variable; M is a sufficiently large positive constant; When two electric vehicles share a charging pile, the charging time period of electric vehicle i (t S,i ,t E,i ) and the charging time period of electric vehicle l (t S,l ,t E,l ) These two non-intersecting constraints should take effect, from which we can deduce:

6. The method for evaluating the peak load shaving response capability of electric vehicles at charging stations considering the vehicle-pile matching relationship according to claim 4 is characterized in that: Design the power constraints of the charging pile and the charging time constraints of the electric vehicle, including: The actual charging power of an electric vehicle is determined by the rated output capacity of the matching charging pile and should not exceed the vehicle's own power tolerance limit: Where: P c,j is the maximum output power of charging pile j; N c is the total number of all available charging piles after the scheduling time node t1; max,i N is the maximum charging power allowed for electric vehicle i; v is the number of electric vehicles to be dispatched; The charging period for electric vehicles must be strictly limited to the service time of the equipment. That is, the charging start time must not be earlier than the charging pile activation time, and the charging end time must not be later than the charging pile deactivation time:

7. The method for evaluating the peak load shaving response capability of electric vehicles at charging stations considering the vehicle-pile matching relationship according to claim 1, characterized in that: Based on the electric vehicle load model and the future deployable charging pile model, the matching constraints between electric vehicles and charging piles, the non-overlapping charging period constraints for multiple vehicles, the power constraints of charging piles, and the charging time constraints for electric vehicles are considered. With the primary goal of fully ensuring users' charging needs, combined with reducing charging costs, an electric vehicle optimization scheduling model is established and solved to obtain the optimal scheduling results, including: Electric vehicle scheduling adopts a multi-objective optimization framework, giving priority to ensuring the satisfaction of charging demand, considering the reduction of total cost, and transforming the multi-objective model into a single-objective model through the weighted coefficient method. The objective function of the electric vehicle optimization scheduling model is: Among them, f is the comprehensive optimization objective function; w1 and w2 are weight coefficients, satisfying w1>>w2; r is the slope function; E i is the total capacity of the power battery pack in electric vehicle i; SOC S,i is the current charge level of electric vehicle i; SOC E,i is the charging termination target value of electric vehicle i; t S,i is the time when electric vehicle i starts charging; t E,i is the end charging time of electric vehicle i; t is time; c t is the electricity price at time t; P c,j is the maximum output power of charging pile j.

8. The method for evaluating the peak load shaving response capability of electric vehicles at charging stations considering the vehicle-pile matching relationship according to claim 1, characterized in that: Based on the electric vehicle load model and the future deployable charging pile model, the matching constraints between electric vehicles and charging piles, the non-overlapping constraints of multiple vehicle charging periods, the power constraints of charging piles, and the charging time constraints of electric vehicles are considered. With the primary goal of fully ensuring users' charging needs, combined with reducing charging costs, an optimal scheduling model for electric vehicles is established and solved, and the optimal scheduling results are obtained, including: When solving the electric vehicle optimization scheduling model, the matching constraints between electric vehicles and charging piles must be followed. In addition, vehicle charging operations must be completed within the user-specified allowed charging time period, that is, the actual charging period must be completely included in the allowed time period: T S,i ≤t S,i ≤t E,i ≤T E,i ,i=1,2,…,N v The total amount of charging should not exceed the required value preset by the user: P v,i (t E,i -t S,i )≤E i (SOC E,i -SOC S,i ),i=1,2,…,N v Among them, E i is the total capacity of the power battery pack in electric vehicle i; SOC S,i is the current charge level of electric vehicle i; SOC E,i is the charging termination target value of electric vehicle i; P v,i is the actual charging power of electric vehicle i, which is determined by the power of the corresponding charging pile that matches the electric vehicle. Its expression is as follows: The end time of electric vehicle charging must be later than the start time: 0≤t S,i ≤t E,i ,i=1,2,…,N v Among them, t S,i is the time when electric vehicle i starts charging; t E,i The moment of final charging for the electric car i.

9. The method for evaluating the peak load shaving response capability of electric vehicles at charging stations considering the vehicle-pile matching relationship according to claim 1, characterized in that: Based on the optimal scheduling results, considering the electric vehicle load translation constraints under different peak shaving response capacities, a charging station peak shaving response capacity evaluation model is established. The relationship between the response amount and the response cost is obtained by solving the charging station peak shaving response capacity evaluation model, including: According to the charging start time and end time of each electric vehicle in the optimal scheduling result, the charging load curve of the electric vehicles at the charging station is obtained as shown below: Among them, P t is the load of the charging station at time t; S,i is the time when electric vehicle i starts charging; t E,i The charging end time of electric vehicle i; P v,i is the actual charging power of electric vehicle i; Considering the electric vehicle load shift constraints under different peak shaving response capacities: the load P of the charging station at time t is calculated as follows: t : For all t S,i and t E,i Sort from small to large and store them in an array. Use variable m to represent the sequence number of the elements in the array, and record the elements in the array as t. l,m ; Establish a set Γ to store the electric vehicle loads that need to be accumulated; Let m = 1, then t l,m Corresponding to the variable t at the start time of charging of an electric vehicle S,i , the electric vehicle load P v,i Add to the set Γ; Let m=m+1, calculate the charging station in the time period [t l,m-1 ,t l,m ) is: If t l,m Corresponding to a variable t representing the starting time of electric vehicle charging S,i , then the load P v,i Add to the set Γ; otherwise, t l,m is a variable t corresponding to the time when the electric vehicle charging ends E,i , then remove the element P from the set Γ v,i ; If t l,m If it is the last element in the array, the charging load curve calculation is completed, otherwise the charging station is recalculated in the period [t l,m-1 ,t l,m ) load.

10. The method for evaluating the peak load shaving response capability of electric vehicles at charging stations considering the vehicle-pile matching relationship according to claim 1, characterized in that: Based on the optimal scheduling results and considering the electric vehicle load shift constraints under different peak shaving response capacities, a charging station peak shaving response capacity evaluation model is established. The relationship between the response amount and the response cost is obtained by solving the charging station peak shaving response capacity evaluation model, including: According to the matching relationship matrix X between electric vehicles and charging piles, the available time period T for each charging pile after scheduling can be obtained: c,j : Among them, T cS,j is the adjustable start time for matching charging pile j; T cE,j To match the end time of the charging pile j; t S,i is the time when electric vehicle i starts charging; t E,i The moment of charging the electric car i ends; Changing the charging period of electric vehicles as a way for electric vehicles to participate in peak shaving response, sorting the charging loads of each electric vehicle in the peak shaving response period, selecting electric vehicles to participate in peak shaving response in the order of gradually increasing peak shaving response amount, and rescheduling the electric vehicles; The optimal scheduling considering peak shaving response not only includes the constraints of the electric vehicle optimal scheduling model, but also includes the constraint that the selected electric vehicles are not charged during the peak shaving response period, which can be expressed as follows: (t S,i -t Ed )(t E,i -t Sd )≥0 Among them, t Sd and t Ed are the start and end times of the peak clipping response respectively; The above formula is converted into mixed integer linear form by the big M method as follows: Among them, y i is an auxiliary 0-1 variable; The cost of peak load response at a charging station consists of two parts: the change in charging electricity charges and the compensation for the reduction in user charging. The following formula is used to measure the cost of demand response at a charging station: Where C is the peak shaving response cost of the charging station; c e,t is the time-of-use electricity price; c m,t is the charging price of the charging pile during the peak electricity price period; t o,S,i and t o,E,i are the charging start and end time of electric vehicle i before peak shaving response; P v,i is the actual charging power of electric vehicle i; N v is the number of electric vehicles to be dispatched; The peak shaving response amount is gradually increased, the above optimization problems are solved in sequence, and the corresponding charging station response costs are calculated, thereby establishing a relationship between the peak shaving response amount and the response cost of the charging station.

Images