Dynamic two-layer optimization scheduling method for electrified transportation-distribution network

Abstract

The present invention relates to the field of distribution network technology, and specifically to a dynamic two-layer optimization scheduling method for an electrified transportation-distribution network, comprising: constructing an objective function of a distribution network optimization scheduling model with the goal of minimizing the sum of active network losses of the distribution network in all time periods within the scheduling time; constructing constraints for the distribution network optimization scheduling model; constructing dynamic equilibrium constraints in the transportation network optimization scheduling model based on the mixed DUE state of the transportation network under an equilibrium path dispatch rate; constructing dynamic equilibrium differential variational inequalities in the transportation network optimization scheduling model; and constructing a dynamic two-layer optimization scheduling model for an electrified transportation-distribution network with the distribution network optimization scheduling model as the upper model and the transportation network scheduling optimization model as the lower model. The present invention controls the temporal and spatial distribution of charging electricity prices and thus controls the comprehensive travel cost of users to achieve orderly charging, optimize traffic flow distribution, effectively reduce distribution network losses, fully utilize the flexibility resources of the dual networks, and smooth out load fluctuations.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution networks, and in particular to a dynamic double-layer optimization scheduling method for an electrified transportation-distribution network. Background Art

[0002] As global attention to sustainable development and environmental protection continues to grow, electrified transportation, as a key means of reducing carbon emissions in the transportation sector, is experiencing rapid growth. At the same time, distribution networks, as a key link in power supply, are facing new challenges and opportunities. The widespread adoption of electrified transportation, such as electric vehicles, has profoundly impacted the operation and planning of distribution networks. Consequently, constructing electrified transportation-distribution network operation models to achieve coordinated optimization has become a research hotspot in the electrical and transportation fields.

[0003] Existing electrified transportation network-distribution network operation models mostly use single-period, long-time-scale static distribution methods or semi-dynamic distribution methods for analysis. They do not consider the impact of the time characteristics of traffic flow and its dynamic changes on the distribution network, and do not consider the comprehensive impact of the characteristics of different spatial areas in urban areas and the spatial characteristics of the distribution network on the operation of the dual networks. It is difficult to meet the needs of transportation network-distribution network operation analysis with flexible resources. Summary of the Invention

[0004] To this end, the technical problem to be solved by the present invention is to overcome the problem that the existing technology does not take into account the impact of the time characteristics of traffic flow and its dynamic changes on the distribution network, and does not take into account the comprehensive impact of the characteristics of different spatial areas in urban areas and the spatial area characteristics of the distribution network on the operation of the dual networks, making it difficult to meet the needs of traffic network-distribution network operation analysis with flexible resources.

[0005] To solve the above technical problems, the present invention provides a dynamic two-layer optimization scheduling method for an electrified transportation-distribution network, comprising:

[0006] The entire scheduling time is divided into multiple periods based on the scheduling time interval; each period is divided into multiple sub-periods based on the discrete duration; the proportion of electric vehicle traffic in the mixed traffic flow on each wireless charging section in each sub-period is preset to be fixed, and the temporal and spatial distribution model of electric vehicle charging load is constructed based on the temporal and spatial distribution of traffic flow;

[0007] Construct the objective function and constraints of the distribution network optimization scheduling model within the scheduling time: Taking the minimum sum of the active network loss of the distribution network in all periods within the scheduling time as the goal, construct the objective function of the distribution network optimization scheduling model; Based on the Distflow branch flow model, combined with the time and space distribution model of the electric vehicle charging load, the SOCR method and the big M method are used to construct the distribution network flow constraints and node voltage constraints that take into account the branch connectivity status after relaxation in the distribution network optimization scheduling model; Based on the connection status and actual sending end of each branch in the distribution network topology, the distribution network topology constraints in the distribution network optimization scheduling model are constructed; Using the big M method, construct the automatic type conversion mechanism model of PV type DG; Construct the energy storage device operation model and the reactive compensation device operation model; Use the automatic type conversion mechanism model of PV type DG, the energy storage device operation model and the reactive compensation device operation model as the operation safety constraints in the distribution network optimization scheduling model;

[0008] A dynamic road network loading model for the transportation network is established; a temporal and spatial distribution model for charging electricity prices is constructed based on the characteristics of urban spatial regions; based on the dynamic road network loading model and the temporal and spatial distribution model for charging electricity prices, expressions for early / late arrival costs, path travel costs, and charging costs are obtained, as well as expressions for the comprehensive path travel costs of fuel vehicles and electric vehicles in the transportation network; based on the expressions for the comprehensive path travel costs of fuel vehicles and electric vehicles in the transportation network, and based on the transportation network being in a mixed DUE state at an equilibrium path dispatch rate, dynamic equilibrium constraints are constructed in the transportation network optimization scheduling model; based on the variational inequality method, dynamic equilibrium differential variational inequalities are constructed in the transportation network optimization scheduling model;

[0009] With the distribution network optimization scheduling model as the upper model and the transportation network scheduling optimization model as the lower model, a dynamic two-layer optimization scheduling model for electrified transportation and distribution network is constructed.

[0010] Initialize the transportation network and distribution network, reconstruct the distribution network within the scheduling time based on the dynamic two-layer optimization scheduling model of electrified transportation and distribution network, obtain the time and space distribution of charging electricity prices, schedule the time and space distribution of charging electricity prices in the transportation network, and output the operating status of the transportation network after optimized scheduling.

[0011] Preferably, the optimization process of the electrified transportation-distribution network dynamic two-layer optimization scheduling model includes:

[0012] Initialize the transportation network and distribution network, obtain the transportation network operation status and congestion data, the distribution network operation status and congestion data, obtain the time and space distribution of traffic flow, and then obtain the time and space distribution of electric vehicle charging load;

[0013] The spatial and temporal distribution of electric vehicle charging, the OD demand of fuel vehicles and electric vehicles in the transportation network, the load factor, the irradiance factor, and the topology of the distribution network are input into the distribution network optimization scheduling model, and the distribution network reconstruction strategy, active and reactive power optimization strategy, photovoltaic output, and the temporal and spatial distribution of charging electricity prices are output;

[0014] The temporal and spatial distribution of charging electricity prices is input into the transportation network optimization scheduling model. The fixed point algorithm is used to schedule the temporal and spatial distribution of charging electricity prices based on the dynamic equilibrium differential variational inequality and combined with the dynamic equilibrium constraints of the transportation network. The equilibrium path dispatch rate is solved and the operation status of the transportation network after optimized scheduling is finally output.

[0015] Preferably, the proportion of electric vehicle traffic in the mixed traffic flow traveling on each wireless charging section in each preset sub-period is fixed. In combination with the temporal and spatial distribution of traffic flow, constructing the temporal and spatial distribution model of electric vehicle charging load includes:

[0016] Assuming that the proportion of electric vehicles in the mixed traffic flow on each wireless charging section in each sub-period is fixed, the active charging load and reactive charging load of electric vehicles provided by the distribution network node a are linearly related to the average traffic flow on the charging circuit powered by the distribution network node a. Therefore, the expressions for the active charging load and reactive charging load of electric vehicles provided by the distribution network node a in sub-period t are:

[0017]

[0018] in, represents the active charging load of electric vehicles provided by distribution network node a in sub-period t; represents the reactive charging load of electric vehicles provided by distribution network node a in sub-period t; ε P Indicates the flow-active load conversion coefficient; ε Q represents the flow-reactive load conversion coefficient; Γ(a) represents the set of wireless charging sections powered by the distribution network node; represents the wireless charging section i powered by the distribution network node a e The average traffic flow of t is the sub-period. N T Indicates the number of sub-periods in a period;

[0019] Based on the number of sub-periods in period T, the expressions for the active charging load and reactive charging load of electric vehicles provided by distribution network node a in period T are obtained as follows:

[0020]

[0021] in, represents the active charging load of electric vehicles provided by distribution network node a during time period T; represents the reactive charging load of electric vehicles provided by the distribution network node a during time period T; T represents the time period, T={T1,T2,...,T L}, L represents the number of time periods in the entire scheduling time;

[0022] Based on the expressions of the active charging load and reactive charging load of electric vehicles provided by each distribution network node in each time period, a temporal and spatial distribution model of electric vehicle charging load is constructed.

[0023] Preferably, the objective function of the distribution network optimization scheduling model is expressed as:

[0024]

[0025] Where T represents the time period, T={T1,T2,...,T L}, L represents the number of time periods in the entire scheduling time; Ω represents the set of branches in the distribution network; k represents a branch; a and b represent the starting node and the end node of the k branch; r k Indicates the resistance of branch k:a~b; Represents the square of the current flowing through branch k:a~b during time period T.

[0026] Preferably, the relaxed distribution network power flow constraints, node voltage constraints, and distribution network topology constraints taking into account branch connectivity states include:

[0027] The expression of the distribution network power flow constraint after relaxation taking into account the branch connectivity state is:

[0028]

[0029] in, It represents the net injected active power of node b during period T, and its expression is: represents the active load of node b during period T, represents the active charging load of the electric vehicle provided by node b during period T; represents the active power of the photovoltaic power generation system connected to node b during time period T; It is represented by the charging and discharging power of the energy storage device connected to node b during time period T; It represents the net injected reactive power of node b during period T, and its expression is: represents the reactive load of node b in the distribution network during time period T; represents the reactive charging load of electric vehicles provided by node b during period T; represents the reactive power of the photovoltaic power generation system connected to node b during time period T; represents the reactive power output of the reactive power compensation device connected to node b during time period T; They represent the active power of branches k and m in the distribution network during time period T respectively; represents the reactive power of branches k and m in the distribution network during time period T; r k Indicates the resistance of branch k:a~b; represents the square of the current flowing through branch k:a~b during time period T; x k represents the reactance on branch k:a~b; Sent(b) represents the set of receiving-end nodes b with sending-end branches in the distribution network; Rec(b) represents the set of sending-end nodes b with receiving-end branches in the distribution network; Ω represents the set of branches in the distribution network; represents a binary variable indicating the connection status of branch k:a~b; M′ represents an additional variable; represents the square of the voltage on node a during time period T; represents the square of the voltage on node b during time period T; T represents time period, T={T1, T2, ..., T L}, L represents the number of time periods in the entire scheduling time;

[0030] The expression of the node voltage constraint is:

[0031]

[0032] v0=1

[0033] Among them, V max Indicates the maximum voltage amplitude of branch η:a~b; V min represents the minimum value of the voltage amplitude of branch η:a~b; Φ represents the node set in the distribution network; v0 represents the voltage of the main substation node;

[0034] The expression of the distribution network topology constraint is:

[0035]

[0036]

[0037] in, Represents a binary variable indicating the connection status of branch k:a~b during time period T. If branch k:a~b is connected, set Otherwise set N is a binary variable indicating the connection status of branch m:c~a during time period T; BRepresents the number of nodes; Rec(a) represents the set of sending-end nodes a with receiving-end branches in the distribution network; Sent(a) represents the set of receiving-end nodes a with sending-end branches in the distribution network; Rec(0) represents the combination of sending-end nodes 0 with receiving-end branches in the distribution network; Ω represents the set of branches in the distribution network; Φ represents the set of nodes in the distribution network; It represents a binary variable indicating the actual sending end of branch k:a~b during time period T. Indicates the binary opposite variable indicating the actual sending end of branch k:a~b during time period T. If the actual sending end of branch k:a~b is a node, then otherwise T represents a time period, T+ΔT represents a time period adjacent to time period T, T={T1, T2, ..., T L}, L represents the number of time periods in the entire scheduling time; ΔT represents the scheduling time interval; n SA Indicates the maximum number of switching actions allowed in the distribution network.

[0038] Preferably, the operational safety constraints include:

[0039] The expression of the automatic type conversion mechanism model of the PV type DG is:

[0040]

[0041] in, A binary variable indicating that the reactive output of the PV-type DG connected to node a reaches its maximum value at time period T. If the reactive output of the PV-type DG connected to node a reaches its maximum value at time period T, then otherwise A binary variable indicating that the reactive output of the PV-type DG connected to node a reaches its minimum value during period T. If the reactive output of the PV-type DG connected to node a reaches its minimum value during period T, then otherwise It represents the maximum reactive output of the PV-type DG connected to node a during time period T; It represents the minimum reactive output of the PV-type DG connected to node a during time period T; represents the reactive power of the photovoltaic power generation system connected to node b during time period T; M * represents a sufficiently large positive number; represents the voltage amplitude of node a during time period T; represents the voltage setting value of node a during time period T; T represents time period, T={T1, T2, ..., T L}, L represents the number of time periods in the entire scheduling time;

[0042] The expression of the energy storage device operation model is:

[0043]

[0044] Among them, E bat,T Indicates the amount of energy stored in the energy storage device during time period T; P ch,T 、P dis,T Respectively represent the charging and discharging power of the energy storage device during time period T; η ch ,η dis Respectively represent the charging and discharging efficiency of the energy storage device; D ch,T 、D dis,T The charging and discharging states of the energy storage device at time period T, D ch,T ∈{0,1},D dis,T ∈{0,1}; They represent the upper limits of charging and discharging power of the energy storage device during time period T respectively; represents the upper limit of the energy storage device; T represents the time period, T={T1,T2,...,T L}, L represents the number of time periods in the entire scheduling time;

[0045] The expression of the reactive power compensation device operation model is:

[0046]

[0047] Among them, Q SVC,T Indicates the reactive output of the reactive compensation device during time period T; Indicates the reactive output upper limit of the reactive compensation device.

[0048] Preferably, the dynamic road network loading model of the transportation network includes: a dynamic point-queue model of the originating node, a road segment model, an intersection model, and a path travel time expression;

[0049] The expression of the originating node dynamic point-queue model is:

[0050]

[0051] Among them, q o (t+Δt) represents the point queue length of the originating node in the sub-period (t+Δt); q o (t) represents the queue length of the originating node o in sub-period t; t+Δt represents the next sub-period adjacent to sub-period t; Δt represents the discrete duration; Π o represents the path set of the originating node o; h p (t) represents the departure rate of path p in sub-period t; D o (t) represents the flow of leaving the queue corresponding to the originating node in sub-period t; Sj (t) represents the supply vehicle flow of the road segment j connected to the originating node o at sub-period t; M j represents a sufficiently large positive number greater than the supply traffic flow of road segment j;

[0052] The adjacent road segments i and j passing through the intersection J are preset to belong to the path p, and the path p belongs to the path set Π of the originating node o. o , then the expression of the road section model is:

[0053]

[0054] Among them, D i (t) represents the required traffic flow of road section i in sub-period t; S j (t) represents the supply vehicle flow of the road segment j connected to the origin node o at sub-period t; Indicates the cumulative number of vehicles entering and leaving road section i in sub-period t; represents the cumulative number of vehicles entering and exiting road section j during sub-period t; represents the incoming traffic volume of road section i at sub-period t; G represents the outgoing traffic flow of road section j in sub-period t; i represents the capacity of road section i; G j represents the capacity of road section j; L i Indicates the length of road section i; L j represents the length of road segment j; represents the wave velocity of the positive wave in section i; represents the wave velocity of the reverse wave on section j; represents the congestion density of road section j; x i (t) represents the average traffic flow of road section i; Δt represents the discrete time length;

[0055] The expression of the intersection model is:

[0056]

[0057]

[0058] Among them, τ i (t) represents the entry time of the vehicle that exits road section i at sub-period t; represents the driving time τ i The cumulative number of vehicles leaving road section i at time (t); represents the percentage of traffic flow belonging to path p on road segment j at sub-period t; represents the traffic volume that passes through the adjacent road segments i and j belonging to path p at the same time in sub-period t; represents the incoming traffic volume of road section j in sub-period t; ij (t) represents the proportion of traffic leaving segment i that enters segment j during sub-period t; represents the driving time τ i The percentage of traffic flow on segment i belonging to path p at time (t); A J (t) represents the traffic distribution matrix of intersection J in sub-period t; represents the outgoing traffic flow of road section i in sub-period t; represents the set of road sections exiting intersection J; represents the set of road sections entering intersection J; represents the outgoing traffic flow of road section i at time period T;

[0059] The path travel time expression is:

[0060]

[0061] Among them, λ i (t) represents the exit time of the vehicle that enters road section i at sub-period t; represents the cumulative number of vehicles entering road section i during sub-period t; represents the departure time λ i The cumulative number of vehicles leaving road section i at time (t); D p (t,h) represents the route departure time at sub-period t and under the vehicle departure rate h; Indicates function nesting; λ o (t) represents the time to leave the queue of the originating node o; λ K (t) represents the time of leaving the queue of node K; g Indicates that it also includes ordinary road section i g and wireless charging section i e A collection of fuel car paths.

[0062] Preferably, the construction of a time-space distribution model of charging electricity prices based on urban spatial regional characteristics includes:

[0063] Based on the spatial characteristics of the city, the city is divided into residential areas, industrial areas, and commercial areas, and a time-space distribution model of charging electricity prices is constructed. The expressions are:

[0064]

[0065] in, They represent the charging electricity prices for residential, industrial, and commercial areas on path p supplied by node a in the distribution network during sub-period t. They represent the basic electricity prices for residential areas, industrial areas, and commercial areas at sub-period t respectively; Respectively represent the service fee coefficients of residential area, industrial area and commercial area in sub-period t; ω t A coefficient indicating the total number of charging points in the area; They represent the service fees of residential area, industrial area and commercial area in sub-period t respectively.

[0066] Preferably, the expressions for early / late arrival cost, path travel cost, and charging cost are obtained based on the dynamic road network loading model of the transportation network and the time-space distribution model of the charging electricity price, and the expressions for the comprehensive path travel cost of fuel vehicles and electric vehicles in the transportation network include:

[0067] The advance / delay costs The expression is:

[0068]

[0069] The path travel cost The expression is:

[0070]

[0071] The charging cost The expression is:

[0072]

[0073] in, Indicates the target time for the vehicle to complete the OD pair (i, j). The OD pair (i, j) indicates that the starting point of the path is i and the end point is j. The OD pair contains multiple paths, forming a path set π ij ;δ represents the travel time cost coefficient;ζ represents the charging cost coefficient;D p (t,h) represents the time of departure from the path at sub-period t and vehicle departure rate h; a∈p represents the node connected to each wireless charging section on path p by the distribution network node a; γ a =γ a (T), represents the charging electricity price at time period T; represents the active charging load of electric vehicles provided by the distribution network node a in the sub-period t; Δt represents the discrete time length; t represents the sub-period, t∈T; T represents the period;

[0074] The comprehensive travel cost of the fuel vehicle path Ψ g,p The expression of (t,h) is:

[0075]

[0076] The comprehensive travel cost of the electric vehicle is Ψ e,p The expression of (t,h) is:

[0077]

[0078] Among them, Π g Indicates that it also includes ordinary road section i g and wireless charging section i e The set of fuel vehicle paths; e Indicates that only wireless charging section i is included e A collection of electric vehicle paths.

[0079] Preferably, the expression of the dynamic equilibrium constraint condition and the dynamic equilibrium differential variational inequality include:

[0080] Based on the traffic network in the equilibrium path departure rate In the mixed DUE state, the non-zero equilibrium path dispatch rate of any type of car in the DUE state has the same minimum comprehensive travel cost, and the equilibrium path dispatch rate whose comprehensive travel cost exceeds the minimum comprehensive travel cost is 0. Then the expression of the dynamic equilibrium constraint condition is:

[0081]

[0082] in, represents the set of feasible equilibrium path departure rates in time period T; h p (t) represents the path departure rate of path p in sub-period t; ij represents the path set of OD pair (i, j); represents the OD traffic demand for (i, j); represents the set of OD pairs in the transportation network; represents the equilibrium path departure rate of path p; g,p (t,h * ) represents the equilibrium path departure rate h in sub-period t * The comprehensive travel cost of the fuel vehicle path under e,p (t,h * ,γ) represents the equilibrium path departure rate h in sub-period t * The comprehensive travel cost of the electric vehicle under the path; γ represents the charging electricity price; They represent the minimum comprehensive travel costs of fuel vehicles and electric vehicles on all the above paths respectively; Respectively represent the path sets of fuel vehicles and electric vehicles on the (o, d) pair; Λ g , Λ e Represent the OD demand pairs of fuel vehicles and electric vehicles respectively;

[0083] The dynamic equilibrium differential variational inequality is:

[0084]

[0085] Among them, p (t,h * ,γ)=Ψ g,p (t,h * )+Ψ e,p (t,h * ,γ) represents the user’s comprehensive travel cost; represents the equilibrium path departure rate of path p in sub-period t.

[0086] The above technical solution of the present invention has the following beneficial effects compared with the prior art:

[0087] (1) The present invention discloses a dynamic two-layer optimization scheduling method for electrified transportation and distribution networks. Based on the spatial regional characteristics of the distribution network, the method changes the distribution network topology by controlling the distribution network tie switches and section switches, reconstructs the distribution network, and considers photovoltaic power generation, reactive power compensation devices, and energy storage devices to construct a distribution network optimization scheduling model to reduce distribution network losses and obtain the temporal and spatial distribution of charging electricity prices.

[0088] (2) The present invention describes a dynamic two-layer optimization scheduling method for an electrified transportation-distribution network. Considering that the dynamic flow distribution calculation model can accurately describe the temporal and spatial evolution of traffic flow in a short period of time, a dynamic road network loading model of the transportation network is constructed based on the dynamic flow distribution calculation model to obtain the comprehensive path cost of fuel vehicles and electric vehicles in the transportation network. The transportation network optimization scheduling model takes the comprehensive cost of users as the target, and when the transportation network is in a mixed DUE state under the equilibrium path departure rate, that is, under the equilibrium path departure rate, both fuel vehicles and electric vehicles are in the DUE state. At this time, all users with the same travel purpose in the transportation network have the same comprehensive travel cost. The comprehensive travel cost of each user is minimized, and no user can obtain a lower cost by changing their route choice. Therefore, the traffic network optimization scheduling model solves the problem of time-space distribution of traffic flow with the user's comprehensive cost as the target, and converts it into a differential variational inequality problem to solve the problem of equilibrium path dispatch rate, so that the operation state of the traffic network reaches the optimal equilibrium state and the choice of each user is optimal; considering the characteristics of different spatial areas in the city, the city is divided into residential areas, industrial areas and commercial areas, and a time-space distribution model of charging electricity price is constructed. Combined with the time-space distribution of traffic flow under the optimal equilibrium state, the charging electricity price can be reasonably adjusted;

[0089] (3) The present invention provides a dynamic two-layer optimization scheduling method for electrified transportation-distribution network, which adopts a two-layer structure. The upper layer constructs a distribution network optimization scheduling model, performs dynamic reconstruction of the distribution network, and actively regulates the distribution network resources to make the distribution network resources fit the distribution of electric vehicle charging load; the lower layer constructs a transportation network optimization scheduling model, and uses the electric vehicle charging load time and space distribution model and the charging electricity price time and space distribution model, taking into account the regional characteristics of different urban areas, to achieve orderly guidance of electric vehicle charging; the two-layer model can effectively reduce the distribution network loss, make full use of the flexibility resources of the two networks, and smooth out load fluctuations, which meets the operation analysis requirements of the transportation network-distribution network containing flexibility resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below based on specific embodiments of the present invention in conjunction with the accompanying drawings, wherein:

[0091] Figure 1 It is a schematic diagram of the electrified transportation-distribution network joint operation model;

[0092] Figure 2 This is a flow chart of a dynamic two-layer optimization scheduling method for electrified transportation-distribution network provided by the present invention;

[0093] Figure 3 It is a schematic diagram of the modified Nguyen-Baran&Wu 33-node system;

[0094] Figure 4 It is a schematic diagram of the optimized distribution network topology;

[0095] Figure 5 It is a line chart of the operating costs before and after optimization;

[0096] Figure 6 It is a line graph of the power of different energy storage systems;

[0097] Figure 7 It is a line graph of the power of different energy storage systems;

[0098] Figure 8 is a schematic diagram of the relative traffic flow distribution before optimization;

[0099] Figure 9 It is a schematic diagram of the relative traffic flow distribution after optimization. DETAILED DESCRIPTION

[0100] The present invention will be further described below with reference to the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it. However, the embodiments are not intended to limit the present invention.

[0101] like Figure 1As shown in Figure 2, the electrified transportation-distribution network joint operation model consists of three parts: transportation network, distribution network and coupling network;

[0102] The traffic network includes electric vehicles and fuel vehicles. In the model, based on road topology and vehicle travel demand, the traffic flow distribution in different time periods and regions is predicted, that is, the time-space distribution of traffic flow, and the time-space distribution of electric vehicle charging load is obtained;

[0103] The distribution network includes distribution network lines, tie switches, section switches, energy storage systems, distributed power sources (PV-type DG), reactive power sources, etc. The reconstruction strategy of the distribution network is obtained based on the initial temporal and spatial distribution data of electric vehicle charging load;

[0104] The transportation network and the distribution network interact through electric vehicle charging facilities and are divided into residential areas, industrial areas, and commercial areas based on the characteristics of urban areas. Through the coupling network, the transportation network provides the distribution network with the temporal and spatial distribution of electric vehicle charging loads, and the distribution network feeds back the operating status of the distribution network to the transportation network.

[0105] Reference Figure 2 As shown, the present invention provides a flow chart of a dynamic two-layer optimization scheduling method for electrified transportation-distribution network; specifically comprising:

[0106] S1: Divide the entire scheduling time into multiple equal-length time periods T according to the scheduling time interval ΔT; where T={T1, T2, ..., T L}, L represents the number of time periods in the entire scheduling time; based on the discrete duration Δt, each time period is divided into multiple sub-time periods t, N T Indicates the number of sub-periods in a period;

[0107] The proportion of electric vehicle traffic in the mixed traffic flow on each wireless charging section in each sub-period is preset to be fixed. Combined with the temporal and spatial distribution of traffic flow, a temporal and spatial distribution model of electric vehicle charging load is constructed, including:

[0108] Assuming that the proportion of electric vehicles in the mixed traffic flow on each wireless charging section in each sub-period is fixed, then c can be considered from a statistical point of view that the active charging load of electric vehicles provided by the distribution network node a is and reactive charging load It is linearly related to the average vehicle flow on the charging circuit powered by the distribution network node a. Then, the expressions of the active charging load and reactive charging load of electric vehicles provided by the distribution network node a in sub-period t are:

[0109]

[0110] in, represents the active charging load of electric vehicles provided by distribution network node a in sub-period t; represents the reactive charging load of electric vehicles provided by distribution network node a in sub-period t; ε P Indicates the flow-active load conversion coefficient; ε Q represents the flow-reactive load conversion coefficient; Γ(a) represents the set of wireless charging sections powered by the distribution network node; represents the wireless charging section i powered by the distribution network node a e The average traffic flow of t is the sub-period. N T Indicates the number of sub-periods in a period;

[0111] Based on the number of sub-periods in period T, the expressions for the active charging load and reactive charging load of electric vehicles provided by distribution network node a in period T are obtained as follows:

[0112]

[0113] in, represents the active charging load of electric vehicles provided by distribution network node a during time period T; represents the reactive charging load of electric vehicles provided by the distribution network node a during time period T; T represents the time period, T={T1,T2,...,T L}, L represents the number of time periods in the entire scheduling time;

[0114] Based on the expressions of the active charging load and reactive charging load of electric vehicles provided by each distribution network node in each time period, a temporal and spatial distribution model of electric vehicle charging load is constructed.

[0115] S2: Construct the objective function and constraints of the distribution network optimization scheduling model within the scheduling time. That is, based on the dynamic DNR problem of the active distribution network, after constructing the original DNR model of the active distribution network, the SOCR method and the large M method are used to transform the original DNR model into a MISCOP problem, and a new DNR model of the active distribution network is obtained as the distribution network optimization scheduling model; the original DNR model of the active distribution network includes the objective function, the distribution network flow constraints taking into account the branch connectivity status, the node voltage constraints, the distribution network topology constraints and the operation safety constraints, specifically including:

[0116] S21: Taking the minimum sum of active network losses of the distribution network in all periods within the dispatching time as the goal, the objective function of the distribution network optimization dispatching model is constructed, and its expression is:

[0117]

[0118] Where T represents the time period, T={T1,T2,...,T L}, L represents the number of time periods in the entire scheduling time; Ω represents the set of branches in the distribution network; k represents a branch; a and b represent the starting node and the end node of the k branch; r k Indicates the resistance of branch k:a~b; It represents the square of the current flowing through branch k:a~b during time period T;

[0119] S22: Based on the Distflow branch power flow model and the temporal and spatial distribution model of electric vehicle charging load, the SOCR method and the Big M method are used to construct the distribution network optimization scheduling model with relaxed distribution network power flow constraints and node voltage constraints that take into account branch connectivity, including:

[0120] Based on the Distflow branch power flow model and the time-space distribution model of electric vehicle charging load, a distribution network power flow constraint that takes into account the branch connectivity status can be constructed. Its expression is:

[0121]

[0122] Since the distribution network power flow constraint taking into account the branch connectivity status contains the multiplication terms of binary variables and continuous variables, as well as binary preconditions, the original DNR model belongs to a mixed integer non-convex nonlinear problem, which is not convenient to solve the original DNR model. Therefore, the SOCR method and the big M method are used to transform the original DNR model into a MISCOP problem, that is, the SOCR method and the big M method are used to transform the distribution network power flow constraint taking into account the branch connectivity status into the relaxed distribution network power flow constraint taking into account the branch connectivity status. Its expression is:

[0123]

[0124] in, It represents the net injected active power of node b during period T, and its expression is: represents the active load of node b during period T, represents the active charging load of the electric vehicle provided by node b during period T; represents the active power of the photovoltaic power generation system connected to node b during time period T; It is represented by the charging and discharging power of the energy storage device connected to node b during time period T; It represents the net injected reactive power of node b during period T, and its expression is: represents the reactive load of node b in the distribution network during time period T; represents the reactive charging load of electric vehicles provided by node b during period T; represents the reactive power of the photovoltaic power generation system connected to node b during time period T; represents the reactive power output of the reactive power compensation device connected to node b during time period T; They represent the active power of branches k and m in the distribution network during time period T respectively; represents the reactive power of branches k and m in the distribution network during time period T; r k Indicates the resistance of branch k:a~b; represents the square of the current flowing through branch k:a~b during time period T; x k represents the reactance on branch k:a~b; Sent(b) represents the set of receiving-end nodes b with sending-end branches in the distribution network; Rec(b) represents the set of sending-end nodes b with receiving-end branches in the distribution network; Ω represents the set of branches in the distribution network; represents a binary variable indicating the connection status of branch k:a~b; M′ represents an additional variable; represents the square of the voltage on node a during time period T; represents the square of the voltage on node b during time period T; T represents time period, T={T1, T2, ..., T L}, L represents the number of time periods in the entire scheduling time;

[0125] In order to estimate the accuracy of the SOCR method, the relative relaxation error (RRE) is defined, which is expressed as:

[0126]

[0127] Among them, Error 1,k Indicates the relative relaxation error corresponding to the branch;

[0128] The expression of the node voltage constraint is:

[0129]

[0130] v0=1

[0131] Among them, V max Indicates the maximum voltage amplitude of branch η:a~b; V min represents the minimum value of the voltage amplitude of branch η:a~b; Φ represents the node set in the distribution network; v0 represents the voltage of the main substation node;

[0132] S23: Based on the connection status and actual sending end of each branch in the distribution network topology, the distribution network topology constraint in the distribution network optimization scheduling model is constructed, and its expression is:

[0133]

[0134]

[0135] in, Represents a binary variable indicating the connection status of branch k:a~b during time period T. If branch k:a~b is connected, set Otherwise set N is a binary variable indicating the connection status of branch m:c~a during time period T; B Represents the number of nodes; Rec(a) represents the set of sending-end nodes a with receiving-end branches in the distribution network; Sent(a) represents the set of receiving-end nodes a with sending-end branches in the distribution network; Rec(0) represents the combination of sending-end nodes 0 with receiving-end branches in the distribution network; Ω represents the set of branches in the distribution network; φ represents the set of nodes in the distribution network; It represents a binary variable indicating the actual sending end of branch k:a~b during time period T. Indicates the binary opposite variable indicating the actual sending end of branch k:a~b during time period T. If the actual sending end of branch k:a~b is a node, then otherwise T represents a time period, T+ΔT represents a time period adjacent to time period T, T={T1, T2, ..., T L}, L represents the number of time periods in the entire scheduling time; ΔT represents the scheduling time interval; n SA Indicates the maximum number of switching actions allowed in the distribution network;

[0136] Among them, the first formula in the distribution network topology constraint satisfies the requirement that if there is no isolated node in the distribution network, there must be no closed loop; the second formula in the distribution network topology constraint satisfies the requirement that the number of branches connected to any node is not equal to 0, that is, any node is not an isolated node; the third and fourth formulas in the distribution network topology constraint meet the requirement that there is no isolated node in the distribution network; the fifth and sixth formulas in the distribution network topology constraint represent auxiliary binary variables and Corresponding constraints, introducing auxiliary binary variables and The purpose is to determine the true flow direction in the branch, so that the phase of the voltage of each node can be recursively calculated step by step from the main substation node; the third formula in the distribution network topology constraint indicates that when branch k is connected, or otherwise The fourth formula in the distribution network topology constraint is used to ensure that all nodes, except the main substation node, have only one branch feeding the power flow. The fifth formula in the distribution network topology constraint means that the true power flow direction in the branch connecting the main substation node 0 and node b must be from node 0 to node b. In other words, the power flow must be fed into the node connected to the main substation node. It is worth noting that the sixth formula in the distribution network topology constraint is based on the assumption that the total generated power of the distributed generation in the distribution network is less than the total load. If this is not the case, this formula can be ignored.

[0137] S24: Constructing operational safety constraints in the distribution network optimization dispatch model, including:

[0138] According to the operation mode and control characteristics of the photovoltaic power station equipped with a voltage-controlled inverter, the photovoltaic power station connected to node a is a PV-type DG, and its active output and voltage amplitude are both set values, that is: and in, represents the active power of the photovoltaic power generation system connected to node b during time period T; represents the voltage amplitude at node b during time period T; represents the active output value of the photovoltaic power generation system connected to node b during time period T; represents the voltage amplitude of the access node b during time period T;

[0139] Generally, in order to keep the bus voltage amplitude at the set value, the photovoltaic power station connected to the grid needs to output a large amount of reactive power; however, the actual output reactive power of photovoltaic power generation is necessarily limited; if the reactive output of the PV type DG exceeds the upper limit or lower limit Its operation mode will be transformed into PQ type DG, and its actual reactive output Q DG,a Will be limited to the limit value; therefore, the big M method is used to construct the automatic type conversion mechanism model of PV type DG, and its expression is:

[0140]

[0141] in, A binary variable indicating that the reactive output of the PV-type DG connected to node a reaches its maximum value at time period T. If the reactive output of the PV-type DG connected to node a reaches its maximum value at time period T, then otherwise A binary variable indicating that the reactive output of the PV-type DG connected to node a reaches its minimum value during period T. If the reactive output of the PV-type DG connected to node a reaches its minimum value during period T, then otherwise It represents the maximum reactive output of the PV-type DG connected to node a during time period T; It represents the minimum reactive output of the PV-type DG connected to node a during time period T; represents the reactive power of the photovoltaic power generation system connected to node b during time period T; M * represents a sufficiently large positive number; represents the voltage amplitude of node a during time period T; represents the voltage setting value of node a during time period T; T represents time period, T={T1, T2, ..., T L}, L represents the number of time periods in the entire scheduling time;

[0142] Construct the energy storage device operation model, and its expression is:

[0143]

[0144] Among them, E bat,T Indicates the amount of energy stored in the energy storage device during time period T; P ch,T 、P dis,T Respectively represent the charging and discharging power of the energy storage device during time period T; η ch ,η dis Respectively represent the charging and discharging efficiency of the energy storage device; D ch,T 、D dis,T The charging and discharging states of the energy storage device at time period T, D ch,T ∈{0,1},D dis,T ∈{0,1}; They represent the upper limits of charging and discharging power of the energy storage device during time period T respectively; represents the upper limit of the energy storage device; T represents the time period, T={T1,T2,...,T L}, L represents the number of time periods in the entire scheduling time;

[0145] The first formula in the energy storage device describes how the energy storage device's charge changes with charging and discharging; the second formula in the energy storage device operation model ensures that the energy storage device's charge remains unchanged throughout the entire scheduling cycle; the third and fourth formulas in the energy storage device operation model limit the energy storage device's charge and discharge power; the fifth formula in the energy storage device operation model ensures that the energy storage device cannot be charged and discharged simultaneously at any time; and the sixth formula in the energy storage device operation model ensures that the energy storage device's charge remains within the range of 20% to 90%, avoiding affecting its service life.

[0146] Construct the reactive power compensation device operation model, and its expression is:

[0147]

[0148] Among them, Q SVC,T Indicates the reactive output of the reactive compensation device during time period T; Indicates the upper limit of reactive output of reactive compensation device;

[0149] The automatic type conversion mechanism model of PV-type DG, the energy storage device operation model and the reactive compensation device operation model are used as operation safety constraints in the distribution network optimization scheduling model.

[0150] S3: The traffic network includes both electric vehicles and fuel vehicles. According to the OD requirements (o, d) (starting point o, end point d) of different users in the traffic network, the set of OD pairs is Π; the vehicles can complete their respective OD pairs through multiple different paths p, and the path set is Π od ; Π g Indicates that it also includes ordinary road section i g and wireless charging section i e The set of fuel vehicle paths; e For the wireless charging section i e The set of electric vehicle paths; o is the set of paths starting from node o; Φ e is the set of distribution network nodes corresponding to the wireless charging section;

[0151] In the case of dynamic traffic allocation, all travelers in the traffic network participate in a non-cooperative Nash game with two degrees of freedom in choosing departure time and travel path. The dynamic traffic flow in the traffic network eventually reaches the DUE state, that is, all users with the same travel purpose have the same comprehensive travel cost. After dividing the time period, the dynamic network loading process of sub-period t in time period T can be described by the dynamic traffic network allocation system after time discretization, which includes:

[0152] S31: Establishing a dynamic road network loading model for the transportation network; wherein the dynamic road network loading model for the transportation network includes: a dynamic point-queue model of the originating node, a road segment model, an intersection model, and a path travel time expression;

[0153] The expression of the originating node dynamic point-queue model is:

[0154]

[0155] Among them, q o (t+Δt) represents the point queue length of the originating node in the sub-period (t+Δt); q o (t) represents the queue length of the originating node o in sub-period t; t+Δt represents the next sub-period adjacent to sub-period t; Δt represents the discrete duration; Π o represents the path set of the originating node o; hp (t) represents the departure rate of path p in sub-period t; D o (t) represents the flow of leaving the queue corresponding to the originating node in sub-period t; S j (t) represents the supply vehicle flow of the road segment j connected to the originating node o at sub-period t; M j represents a sufficiently large positive number greater than the supply traffic flow of road segment j;

[0156] The adjacent road segments i and j passing through the intersection J are preset to belong to the path p, and the path p belongs to the path set Π of the originating node o. o , then the expression of the road section model is:

[0157]

[0158] Among them, D i (t) represents the required traffic flow of road section i in sub-period t; S j (t) represents the supply vehicle flow of the road segment j connected to the origin node o at sub-period t; Indicates the cumulative number of vehicles entering and leaving road section i in sub-period t; represents the cumulative number of vehicles entering and exiting road section j during sub-period t; represents the incoming traffic volume of road section i at sub-period t; G represents the outgoing traffic flow of road section j in sub-period t; i represents the capacity of road section i; G j represents the capacity of road section j; L i Indicates the length of road section i; L j represents the length of road segment j; represents the wave velocity of the positive wave in section i; represents the wave velocity of the reverse wave on section j; represents the congestion density of road section j; x i (t) represents the average traffic flow of road section i; Δt represents the discrete time length;

[0159] The expression of the intersection model is:

[0160]

[0161] Among them, τ i (t) represents the entry time of the vehicle that exits road section i at sub-period t; represents the driving time τ i The cumulative number of vehicles leaving road section i at time (t); represents the percentage of traffic flow belonging to path p on road segment j at sub-period t; represents the traffic volume that passes through the adjacent road segments i and j belonging to path p at the same time in sub-period t; represents the incoming traffic volume of road section j in sub-period t; ij (t) represents the proportion of traffic leaving segment i that enters segment j during sub-period t; represents the driving time τ i The percentage of traffic flow on segment i belonging to path p at time (t); A J (t) represents the traffic distribution matrix of intersection J in sub-period t; represents the outgoing traffic flow of road section i in sub-period t; represents the set of road sections exiting intersection J; represents the set of road sections entering intersection J; represents the outgoing traffic flow of road section i at time period T;

[0162] The path travel time expression is:

[0163]

[0164] Among them, λ i (t) represents the exit time of the vehicle that enters road section i at sub-period t; represents the cumulative number of vehicles entering road section i during sub-period t; represents the departure time λ i The cumulative number of vehicles leaving road section i at time (t); D p (t,h) represents the route departure time at sub-period t and under the vehicle departure rate h; Indicates function nesting; λ o (t) represents the time to leave the queue of the originating node o; λ K (t) represents the time of leaving the queue of node K; g Indicates that it also includes ordinary road section i g and wireless charging section i e The set of fuel vehicle paths;

[0165] S32: Based on the characteristics of urban spatial regions, a temporal and spatial distribution model of charging electricity prices is constructed, including:

[0166] Based on the spatial characteristics of the city, the city is divided into residential areas, industrial areas, and commercial areas, and a time-space distribution model of charging electricity prices is constructed. The expressions are:

[0167]

[0168] in, They represent the charging electricity prices for residential, industrial, and commercial areas on path p supplied by node a in the distribution network during sub-period t. They represent the basic electricity prices for residential areas, industrial areas, and commercial areas at sub-period t respectively; Respectively represent the service fee coefficients of residential area, industrial area and commercial area in sub-period t; ω t A coefficient indicating the total number of charging points in the area; represent the service fees for residential areas, industrial areas, and commercial areas at sub-period t, respectively;

[0169] S33: Based on the dynamic road network loading model of the transportation network and the time-space distribution model of charging electricity prices, expressions for early / delayed arrival costs, path travel costs, and charging costs are obtained, as well as expressions for the comprehensive path travel costs of fuel vehicles and electric vehicles in the transportation network, including:

[0170] The advance / delay costs The expression is:

[0171]

[0172] The path travel cost The expression is:

[0173]

[0174] The charging cost The expression is:

[0175]

[0176] in, Indicates the target time for the vehicle to complete the OD pair (i, j). The OD pair (i, j) indicates that the starting point of the path is i and the end point is j. The OD pair contains multiple paths, forming a path set π ij ;δ represents the travel time cost coefficient;ζ represents the charging cost coefficient;D p (t,h) represents the time of departure from the path at sub-period t and vehicle departure rate h; a∈p represents the node connected to each wireless charging section on path p by the distribution network node a; γ a =γ a (T), represents the charging electricity price at time period T; represents the active charging load of electric vehicles provided by the distribution network node a in the sub-period t; Δt represents the discrete time length; t represents the sub-period, t∈T; T represents the period;

[0177] The comprehensive travel cost of the fuel vehicle path Ψ g,p The expression of (t,h) is:

[0178]

[0179] The comprehensive travel cost of the electric vehicle is Ψ e,p The expression of (t,h) is:

[0180]

[0181] Among them, Π g Indicates that it also includes ordinary road section i g and wireless charging section i e The set of fuel vehicle paths; e Indicates that only wireless charging section i is included e The set of electric vehicle paths;

[0182] S34: Based on the expression of the comprehensive travel cost of fuel vehicles and electric vehicles in the transportation network, and assuming that the transportation network is in a mixed DUE state under the equilibrium path dispatch rate, the dynamic equilibrium constraints in the transportation network optimization scheduling model are constructed, including:

[0183] For a traffic network that contains both fuel vehicles and electric vehicles, if you want to meet the equilibrium path departure rate In the mixed DUE state, it is valid only if and only if the following conditions are met, that is, the conditions must be met: in the DUE state, the non-zero equilibrium path dispatch rate of any type of car has the same minimum comprehensive travel cost, and the equilibrium path dispatch rate whose comprehensive travel cost exceeds the minimum comprehensive travel cost is 0. Then the expression of the dynamic equilibrium constraint condition is:

[0184]

[0185] in, represents the set of feasible equilibrium path departure rates in time period T; h p (t) represents the path departure rate of path p in sub-period t; ij represents the path set of OD pair (i, j); represents the traffic demand of OD pair (i, j); w represents the set of OD pairs in the transportation network; represents the equilibrium path departure rate of path p; g,p (t,h * ) represents the equilibrium path departure rate h in sub-period t * The comprehensive travel cost of the fuel vehicle path under e,p (t,h * ,γ) represents the equilibrium path departure rate h in sub-period t * The comprehensive travel cost of the electric vehicle under the path; γ represents the charging electricity price; They represent the minimum comprehensive travel costs of fuel vehicles and electric vehicles on all the above paths respectively; Respectively represent the path sets of fuel vehicles and electric vehicles on the (o, d) pair; Λ g , Λ e Represent the OD demand pairs of fuel vehicles and electric vehicles respectively;

[0186] In order to obtain the traffic flow distribution within the region, it is necessary to solve its mixed DUE state, that is, to satisfy the above-mentioned dynamic equilibrium constraints. This is equivalent to the following differential variational inequality problem. Therefore, based on the variational inequality method, the dynamic equilibrium differential variational inequality in the traffic network optimization scheduling model can be constructed, and its expression is:

[0187]

[0188] Among them, p (t,h * ,γ)=Ψ g,p (t,h * )+Ψ e,p (t,h * ,γ) represents the user’s comprehensive travel cost; represents the equilibrium path departure rate of path p in sub-period t;

[0189] Furthermore, based on the equivalence of differential variational inequalities and fixed point problems, a fixed point algorithm can be established to solve dynamic equilibrium differential variational inequalities and obtain the fixed point The value of .

[0190] S4: Using the distribution network optimization scheduling model as the upper model and the transportation network optimization scheduling model as the lower model, a dynamic two-layer optimization scheduling model for electrified transportation and distribution networks is constructed;

[0191] S5: The optimization process of the electrified transportation-distribution network dynamic two-layer optimization scheduling model includes:

[0192] Initialize the transportation network and distribution network, obtain the transportation network operation status and congestion data, the distribution network operation status and congestion data, obtain the time and space distribution of traffic flow, and then obtain the time and space distribution of electric vehicle charging load;

[0193] The spatial and temporal distribution of electric vehicle charging, the OD demand of fuel vehicles and electric vehicles in the transportation network, the load factor, the irradiance factor, and the topology of the distribution network are input into the distribution network optimization scheduling model, and the distribution network reconstruction strategy, active and reactive power optimization strategy, photovoltaic output, and the temporal and spatial distribution of charging electricity prices are output;

[0194] The temporal and spatial distribution of charging electricity prices is input into the transportation network optimization scheduling model. The fixed point algorithm is used to schedule the temporal and spatial distribution of charging electricity prices based on the dynamic equilibrium differential variational inequality and combined with the dynamic equilibrium constraints of the transportation network. The equilibrium path dispatch rate is solved and the operation status of the transportation network after optimized scheduling is finally output.

[0195] In order to verify that the electrified transportation-distribution network dynamic two-layer optimization scheduling model designed by the present invention can effectively reduce the distribution network loss and obtain a better dual-network operation state; Figure 3 As shown, the present invention uses a modified Nguyen network—the Baran & Wu 33-node system. The Nguyen network includes both electric and fuel-powered vehicles. The road network consists of 13 nodes, 19 road segments, two OD demand pairs (OD(1,2) and OD(4,2), and 24 paths. The congestion densities for road segments 11, 14, and 16 are 0.06, 0.17, and 0.17 veh / m, respectively, and their lengths are 9, 3, and 3 km, respectively. The subperiod Δt is set to 180 seconds.

[0196] The distribution network includes two photovoltaic power plants, PV1 and PV2, corresponding to nodes 14 and 29, respectively. The two compensation capacitor banks, CB1 and CB2, have a maximum switching capacity of three groups and a maximum switching frequency of three times within a dispatch cycle. Each bank provides a compensation power of 50 kVar, corresponding to nodes 8 and 25, respectively. The two SVCs, SVC1 and SVC2, correspond to nodes 9 and 26, respectively, with an adjustable power range of -300 kVar to 300 kVar. The node voltage constraint range is 0.95 pu to 1.05 pu. The solution was performed using the Mosek algorithm package in Matlab R2018b. The system hardware environment required a 3.2 GHz CPU and 8 GB of RAM.

[0197] The goal is to minimize network losses during the upper distribution network operation cycle. Figures 4 to 7 The upper layer changes the distribution network topology by controlling link switches and segment switches, and regulates more resources for the electric vehicle charging load. For section 11, for distribution network node 17, after the topology is changed, its position is changed from the end of the distribution network to the middle. Through the regulation of flexible resources, the source-load matching conflict problem is effectively alleviated, thereby stabilizing the voltage of the distribution network, reducing network losses, and effectively reducing the operating costs of the dual networks.

[0198] like Figures 8-9As shown in the figure, after lower-level optimization scheduling, the corresponding electric vehicle charging prices were adjusted. The maximum relative traffic flow on section 15 decreased from 8.3% to 0.03%, the maximum relative traffic flow on section 161 increased from 12% to 21%, and the maximum relative traffic flow on section 11 decreased from 9.8% to 9.7%. By changing the distribution of electricity prices and thus affecting users' comprehensive travel costs, electric vehicle traffic can be effectively controlled, alleviating congestion in both the transportation network and the distribution network.

[0199] When the charging load of electric vehicles increases, or the voltage of some nodes in the distribution network decreases, the topology of the upper distribution network is changed, and the electricity price distribution of the lower transportation network is changed, which can optimize the scheduling of dual-network resources and alleviate congestion.

[0200] In the electrified transportation-distribution network joint operation optimization model, both networks have flexible dispatchable resources, but need to be optimized in terms of time and space; the electrified transportation-distribution network dynamic two-layer optimization scheduling model provided by the present invention is as follows Figure 2 As shown in the figure; the initialization part takes the user's comprehensive travel cost as the target, obtains the initial traffic network operation status and congestion data and the distribution network operation status and congestion data, and then extracts the time and space distribution of electric vehicle charging load; the upper distribution network optimization scheduling model takes the distribution network loss as the target, inputs the vehicle network data, including OD demand, load rate, illumination rate, etc., changes the distribution network topology by controlling the distribution network tie switch and section switch, and uses SVC (reactive compensation device) and energy storage device to reduce network loss; the lower traffic network optimization scheduling model further regulates the time and space distribution of transportation network electricity price according to the optimization results of the upper model and the characteristics of the urban area; the city can be roughly divided into residential areas, commercial areas, and industrial areas according to regional characteristics, and the charging price control range is set according to regional characteristics; the user's comprehensive travel cost is taken as the target, including early / delayed arrival cost, path driving cost and charging cost.

[0201] The present invention proposes a dynamic two-layer optimization scheduling model for electrified transportation and distribution networks. The upper layer constructs a dynamic reconstruction model for the distribution network, actively regulates the distribution network resources, and makes the distribution network resources fit the distribution of electric vehicle charging loads. The lower layer constructs a temporal and spatial distribution model of the electric vehicle charging load electricity price, taking into account the regional characteristics of different urban areas, and realizing orderly guidance of electric vehicle charging. The two-layer model can effectively reduce distribution network losses, make full use of the flexibility resources of the dual networks, and smooth out load fluctuations.

[0202] The present invention provides a dynamic two-layer optimization scheduling model for a transportation network and distribution network containing mixed traffic of electric vehicles and fuel vehicles. The model uses a dynamic flow distribution method to provide initial traffic flow distribution data. The upper layer is the dynamic reconstruction problem of the distribution network containing electric vehicle charging loads. The lower layer uses dynamic traffic flow to characterize the temporal and spatial changes of traffic flow. By controlling the temporal and spatial distribution of electricity prices and thus the comprehensive travel cost of users, orderly charging is achieved and traffic flow distribution is optimized.

[0203] Obviously, the above embodiments are merely examples for clarity of explanation and are not intended to limit the implementation methods. Those skilled in the art will appreciate that other variations or modifications can be made based on the above description. It is not necessary and impossible to enumerate all implementation methods here. Obvious variations or modifications arising therefrom remain within the scope of protection of the present invention.

Claims

1. A dynamic two-layer optimization scheduling method for electrified transportation-distribution network, characterized in that: include: According to the scheduling time interval, the entire scheduling time is divided into multiple time periods; Based on discrete duration, each period is divided into multiple sub-periods; The proportion of electric vehicle traffic in the mixed traffic flow on each wireless charging section in each sub-period is preset to be fixed. Combined with the temporal and spatial distribution of traffic flow, a temporal and spatial distribution model of electric vehicle charging load is constructed. Construct the objective function and constraints of the distribution network optimization scheduling model within the scheduling time: Taking the minimum sum of the active network loss of the distribution network in all periods within the scheduling time as the goal, construct the objective function of the distribution network optimization scheduling model; Based on the Distflow branch flow model, combined with the time and space distribution model of the electric vehicle charging load, the SOCR method and the big M method are used to construct the distribution network flow constraints and node voltage constraints that take into account the branch connectivity status after relaxation in the distribution network optimization scheduling model; Based on the connection status and actual sending end of each branch in the distribution network topology, the distribution network topology constraints in the distribution network optimization scheduling model are constructed; Using the big M method, construct the automatic type conversion mechanism model of PV type DG; Construct the energy storage device operation model and the reactive compensation device operation model; Use the automatic type conversion mechanism model of PV type DG, the energy storage device operation model and the reactive compensation device operation model as the operation safety constraints in the distribution network optimization scheduling model; A dynamic road network loading model for the transportation network is established; a temporal and spatial distribution model for charging electricity prices is constructed based on the characteristics of urban spatial regions; based on the dynamic road network loading model and the temporal and spatial distribution model for charging electricity prices, expressions for early / late arrival costs, path travel costs, and charging costs are derived, as well as expressions for the comprehensive path travel costs of fuel vehicles and electric vehicles in the transportation network; based on the expressions for the comprehensive path travel costs of fuel vehicles and electric vehicles in the transportation network, and assuming that the transportation network is in a mixed DUE state at an equilibrium path dispatch rate, dynamic equilibrium constraints are constructed in the transportation network optimization scheduling model; Based on the variational inequality method, the dynamic equilibrium differential variational inequality in the transportation network optimization scheduling model is constructed; With the distribution network optimization scheduling model as the upper model and the transportation network scheduling optimization model as the lower model, a dynamic two-layer optimization scheduling model for electrified transportation and distribution network is constructed. Initialize the transportation network and distribution network, reconstruct the distribution network within the scheduling time based on the dynamic two-layer optimization scheduling model of electrified transportation and distribution network, obtain the time-space distribution of charging electricity prices, and then input the time-space distribution of charging electricity prices into the transportation network optimization scheduling model. Using the fixed point algorithm, based on the dynamic equilibrium differential variational inequality and combined with the dynamic equilibrium constraints of the transportation network, the time-space distribution of charging electricity prices in the transportation network is scheduled, the equilibrium path dispatch rate is solved, and the operating status of the transportation network after optimized scheduling is output.

2. The method for dynamic two-layer optimization scheduling of electrified transportation-distribution network according to claim 1, characterized in that: The optimization process of the electrified transportation-distribution network dynamic two-layer optimization scheduling model includes: Initialize the transportation network and distribution network, obtain the transportation network operation status and congestion data, the distribution network operation status and congestion data, obtain the time and space distribution of traffic flow, and then obtain the time and space distribution of electric vehicle charging load; The spatial and temporal distribution of electric vehicle charging, the OD demand of fuel vehicles and electric vehicles in the transportation network, the load factor, the irradiance factor, and the topology of the distribution network are input into the distribution network optimization scheduling model, and the distribution network reconstruction strategy, active and reactive power optimization strategy, photovoltaic output, and the temporal and spatial distribution of charging electricity prices are output; The temporal and spatial distribution of charging electricity prices is input into the transportation network optimization scheduling model. The fixed point algorithm is used to schedule the temporal and spatial distribution of charging electricity prices based on the dynamic equilibrium differential variational inequality and combined with the dynamic equilibrium constraints of the transportation network. The equilibrium path dispatch rate is solved and the operation status of the transportation network after optimized scheduling is finally output.

3. The method for dynamic two-layer optimization scheduling of electrified transportation-distribution network according to claim 1, characterized in that: The proportion of electric vehicle traffic in the mixed traffic flow on each wireless charging section in each preset sub-period is fixed. Combined with the temporal and spatial distribution of traffic flow, the temporal and spatial distribution model of electric vehicle charging load is constructed, which includes: Assuming that the proportion of electric vehicles in the mixed traffic flow on each wireless charging section in each sub-period is fixed, the active charging load and reactive charging load of electric vehicles provided by the distribution network node a are linearly related to the average traffic flow on the charging circuit powered by the distribution network node a. Therefore, the expressions for the active charging load and reactive charging load of electric vehicles provided by the distribution network node a in sub-period t are: in, represents the active charging load of electric vehicles provided by distribution network node a in sub-period t; represents the reactive charging load of electric vehicles provided by distribution network node a in sub-period t; ε P Indicates the flow-active load conversion coefficient; ε Q represents the flow-reactive load conversion coefficient; Γ(a) represents the set of wireless charging sections powered by the distribution network node; represents the wireless charging section i powered by the distribution network node a e The average traffic flow of t is the sub-period. N T Indicates the number of sub-periods in a period; Based on the number of sub-periods in period T, the expressions for the active charging load and reactive charging load of electric vehicles provided by distribution network node a in period T are obtained as follows: in, represents the active charging load of electric vehicles provided by distribution network node a during time period T; represents the reactive charging load of electric vehicles provided by the distribution network node a during time period T; T represents the time period, T={T1,T2,...,T L }, L represents the number of time periods in the entire scheduling time; Based on the expressions of the active charging load and reactive charging load of electric vehicles provided by each distribution network node in each time period, a temporal and spatial distribution model of electric vehicle charging load is constructed.

4. The method for dynamic two-layer optimization scheduling of electrified transportation-distribution network according to claim 1, characterized in that: The objective function of the distribution network optimization scheduling model is expressed as follows: Where T represents the time period, T={T1,T2,...,T L }, L represents the number of time periods in the entire scheduling time; Ω represents the set of branches in the distribution network; k represents a branch; a and b represent the starting node and the end node of the k branch; r k Indicates the resistance of branch k:a~b; Represents the square of the current flowing through branch k:a~b during time period T.

5. The method for dynamic double-layer optimization scheduling of electrified transportation-distribution network according to claim 1, characterized in that: The relaxed distribution network power flow constraints, node voltage constraints, and distribution network topology constraints that take into account branch connectivity include: The expression of the distribution network power flow constraint after relaxation taking into account the branch connectivity state is: in, It represents the net injected active power of node b during period T, and its expression is: represents the active load of node b during period T, represents the active charging load of the electric vehicle provided by node b during period T; represents the active power of the photovoltaic power generation system connected to node b during time period T; It is represented by the charging and discharging power of the energy storage device connected to node b during time period T; It represents the net injected reactive power of node b during period T, and its expression is: represents the reactive load of node b in the distribution network during time period T; represents the reactive charging load of electric vehicles provided by node b during period T; represents the reactive power of the photovoltaic power generation system connected to node b during time period T; represents the reactive power output of the reactive power compensation device connected to node b during time period T; They represent the active power of branches k and m in the distribution network during time period T respectively; represents the reactive power of branches k and m in the distribution network during time period T; r k Indicates the resistance of branch k:a~b; represents the square of the current flowing through branch k:a~b during time period T; x k represents the reactance on branch k:a~b; Sent(b) represents the set of receiving-end nodes b with sending-end branches in the distribution network; Rec(b) represents the set of sending-end nodes b with receiving-end branches in the distribution network; Ω represents the set of branches in the distribution network; M is a binary variable indicating the connection status of branch k:a~b; ′ Indicates additional variables; represents the square of the voltage on node a during time period T; represents the square of the voltage on node b during time period T; T represents time period, T={T1, T2, ..., T L }, L represents the number of time periods in the entire scheduling time; The expression of the node voltage constraint is: v0＝1 Among them, V max Indicates the maximum voltage amplitude of branch η:a~b; V min represents the minimum value of the voltage amplitude of branch η:a~b; Φ represents the node set in the distribution network; v0 represents the voltage of the main substation node; The expression of the distribution network topology constraint is: in, Represents a binary variable indicating the connection status of branch k:a~b during time period T. If branch k:a~b is connected, set Otherwise set N is a binary variable indicating the connection status of branch m:c~a during time period T; B Represents the number of nodes; Rec(a) represents the set of sending-end nodes a with receiving-end branches in the distribution network; Sent(a) represents the set of receiving-end nodes a with sending-end branches in the distribution network; Rec(0) represents the combination of sending-end nodes 0 with receiving-end branches in the distribution network; Ω represents the set of branches in the distribution network; Φ represents the set of nodes in the distribution network; It represents a binary variable indicating the actual sending end of branch k:a~b during time period T. Indicates the binary opposite variable indicating the actual sending end of branch k:a~b during time period T. If the actual sending end of branch k:a~b is a node, then otherwise T represents a time period, T+ΔT represents a time period adjacent to time period T, T={T1, T2, ..., T L }, L represents the number of time periods in the entire scheduling time; ΔT represents the scheduling time interval; n SA Indicates the maximum number of switching actions allowed in the distribution network.

6. The method for dynamic double-layer optimization scheduling of electrified transportation-distribution network according to claim 1, characterized in that: The operational safety constraints include: The expression of the automatic type conversion mechanism model of the PV type DG is: in, A binary variable indicating that the reactive output of the PV-type DG connected to node a reaches its maximum value at time period T. If the reactive output of the PV-type DG connected to node a reaches its maximum value at time period T, then otherwise A binary variable indicating that the reactive output of the PV-type DG connected to node a reaches its minimum value during period T. If the reactive output of the PV-type DG connected to node a reaches its minimum value during period T, then otherwise It represents the maximum reactive output of the PV-type DG connected to node a during time period T; It represents the minimum reactive output of the PV-type DG connected to node a during time period T; represents the reactive power of the photovoltaic power generation system connected to node b during time period T; M * represents a sufficiently large positive number; represents the voltage amplitude of node a during time period T; represents the voltage setting value of node a during time period T; T represents time period, T={T1, T2, ..., T L }, L represents the number of time periods in the entire scheduling time; The expression of the energy storage device operation model is: Among them, E bat,T Indicates the amount of energy stored in the energy storage device during time period T; P ch,T 、P dis,T Respectively represent the charging and discharging power of the energy storage device during time period T; η ch ,η dis Respectively represent the charging and discharging efficiency of the energy storage device; D ch,T 、D dis,T The charging and discharging states of the energy storage device at time period T, D ch,T ∈{0,1},D dis,T ∈{0,1}; They represent the upper limits of charging and discharging power of the energy storage device during time period T respectively; represents the upper limit of the energy storage device; T represents the time period, T={T1,T2,...,T L }, L represents the number of time periods in the entire scheduling time; The expression of the reactive power compensation device operation model is: Among them, Q SVC,T Indicates the reactive output of the reactive compensation device during time period T; Indicates the reactive output upper limit of the reactive compensation device.

7. The method for dynamic double-layer optimization scheduling of electrified transportation-distribution network according to claim 1, characterized in that: The dynamic road network loading model of the transportation network includes: a dynamic point-queue model of the originating node, a road segment model, an intersection model, and a path travel time expression; The expression of the originating node dynamic point-queue model is: Among them, q o (t+Δt) represents the point queue length of the originating node in the sub-period (t+Δt); q o (t) represents the queue length of the originating node o in sub-period t; t+Δt represents the next sub-period adjacent to sub-period t; Δt represents the discrete duration; Π o represents the path set of the originating node o; h p (t) represents the departure rate of path p in sub-period t; D o (t) represents the flow of leaving the queue corresponding to the originating node in sub-period t; S j (t) represents the supply vehicle flow of the road segment j connected to the originating node o at sub-period t; M j represents a sufficiently large positive number greater than the supply traffic flow of road segment j; The adjacent road segments i and j passing through the intersection J are preset to belong to the path p, and the path p belongs to the path set Π of the originating node o. o , then the expression of the road section model is: Among them, D i (t) represents the required traffic flow of road section i in sub-period t; S j (t) represents the supply vehicle flow of the road segment j connected to the origin node o at sub-period t; Indicates the cumulative number of vehicles entering and leaving road section i in sub-period t; represents the cumulative number of vehicles entering and exiting the road section j during sub-period t; f i in (t) represents the traffic flow of road section i in sub-period t; f j out (t) represents the outgoing traffic flow of section j in sub-period t; G i represents the capacity of road section i; G j represents the capacity of road section j; L i Indicates the length of road section i; L j represents the length of road segment j; represents the wave velocity of the positive wave in section i; represents the wave velocity of the reverse wave on section j; represents the congestion density of road section j; x i (t) represents the average traffic flow of road section i; Δt represents the discrete time length; The expression of the intersection model is: Among them, τ i (t) represents the entry time of the vehicle that exits road section i at sub-period t; represents the driving time τ i The cumulative number of vehicles leaving road section i at time (t); represents the percentage of traffic flow belonging to path p on road segment j at sub-period t; represents the traffic volume that passes through the adjacent road segments i and j belonging to path p at the same time in sub-period t; f j in (t) represents the incoming traffic volume of road section j in sub-period t; ij (t) represents the proportion of traffic leaving segment i that enters segment j during sub-period t; represents the driving time τ i The percentage of traffic flow on segment i belonging to path p at time (t); A J (t) represents the traffic distribution matrix of intersection J in sub-period t; f i out (t) represents the outgoing traffic flow of road section i in sub-period t; represents the set of road sections exiting intersection J; represents the set of road sections entering intersection J; f i out (T) represents the outgoing traffic volume of road section i during time period T; The path travel time expression is: Among them, λ i (t) represents the exit time of the vehicle that enters road section i at sub-period t; represents the cumulative number of vehicles entering road section i during sub-period t; represents the departure time λ i The cumulative number of vehicles leaving road section i at time (t); D p (t,h) represents the route departure time at sub-period t and under the vehicle departure rate h; Indicates function nesting; λ o (t) represents the time to leave the queue of the originating node o; λ K (t) represents the time of leaving the queue of node K; g Indicates that it also includes ordinary road section i g and wireless charging section i e A collection of fuel car paths.

8. The method for dynamic double-layer optimization scheduling of electrified transportation-distribution network according to claim 1, characterized in that: The construction of a time-space distribution model of charging electricity prices based on urban spatial regional characteristics includes: Based on the spatial characteristics of the city, the city is divided into residential areas, industrial areas, and commercial areas, and a time-space distribution model of charging electricity prices is constructed. The expressions are: in, They represent the charging electricity prices for residential, industrial, and commercial areas on path p supplied by node a in the distribution network during sub-period t. They represent the basic electricity prices for residential areas, industrial areas, and commercial areas at sub-period t respectively; Respectively represent the service fee coefficients of residential area, industrial area and commercial area in sub-period t; ω t A coefficient indicating the total number of charging points in the area; They represent the service fees of residential area, industrial area and commercial area in sub-period t respectively.

9. The method for dynamic double-layer optimization scheduling of electrified transportation-distribution network according to claim 1, characterized in that: Based on the dynamic road network loading model of the transportation network and the time-space distribution model of charging electricity prices, the expressions for early / delayed arrival cost, path travel cost, and charging cost are obtained, as well as the expression for the comprehensive path travel cost of fuel vehicles and electric vehicles in the transportation network, including: The advance / delay costs The expression is: The path travel cost The expression is: The charging cost The expression is: in, Indicates the target time for the vehicle to complete the OD pair (i, j). The OD pair (i, j) indicates that the starting point of the path is i and the end point is j. The OD pair contains multiple paths, forming a path set π ij ;δ represents the travel time cost coefficient;ζ represents the charging cost coefficient;D p (t,h) represents the time of departure from the path at sub-period t and vehicle departure rate h; a∈p represents the node connected to each wireless charging section on path p by the distribution network node a; γ a =γ a (T), represents the charging electricity price at time period T; represents the active charging load of electric vehicles provided by the distribution network node a in the sub-period t; Δt represents the discrete time length; t represents the sub-period, t∈T; T represents the period; The comprehensive travel cost of the fuel vehicle path Ψ g,p The expression of (t,h) is: The comprehensive travel cost of the electric vehicle is Ψ e,p The expression of (t,h) is: Among them, Π g Indicates that it also includes ordinary road section i g and wireless charging section i e The set of fuel vehicle paths; e Indicates that only wireless charging section i is included e A collection of electric vehicle paths.

10. The method for dynamic double-layer optimization scheduling of electrified transportation-distribution network according to claim 1, characterized in that: The expressions of dynamic equilibrium constraints and dynamic equilibrium differential variational inequalities include: Based on the traffic network in the equilibrium path departure rate In the mixed DUE state, the non-zero equilibrium path dispatch rate of any type of car in the DUE state has the same minimum comprehensive travel cost, and the equilibrium path dispatch rate whose comprehensive travel cost exceeds the minimum comprehensive travel cost is 0. Then the expression of the dynamic equilibrium constraint condition is: in, represents the set of feasible equilibrium path departure rates in time period T; h p (t) represents the path departure rate of path p in sub-period t; ij represents the path set of OD pair (i, j); represents the OD traffic demand for (i, j); represents the set of OD pairs in the transportation network; represents the equilibrium path departure rate of path p; g,p (t,h * ) represents the equilibrium path departure rate h in sub-period t * The comprehensive travel cost of the fuel vehicle path under e,p (t,h * ,γ) represents the equilibrium path departure rate h in sub-period t * The comprehensive travel cost of the electric vehicle under the following path; γ represents the charging electricity price; They represent the minimum comprehensive travel costs of fuel vehicles and electric vehicles on all the above paths respectively; Respectively represent the path sets of fuel vehicles and electric vehicles on the (o, d) pair; Λ g , Λ e Represent the OD demand pairs of fuel vehicles and electric vehicles respectively; The dynamic equilibrium differential variational inequality is: Among them, p (t,h * ,γ)=Ψ g,p (t,h * )+Ψ e,p (t,h * ,γ) represents the user’s comprehensive travel cost; represents the equilibrium path departure rate of path p in sub-period t.

Images