A method for coordinated dispatching of medium and low voltage distribution networks based on flexible resources on the low voltage side

Abstract

The present invention discloses a coordinated dispatching method for medium- and low-voltage distribution networks based on flexible resources on the low-voltage side, including the following steps: first, considering that the travel behavior of a single electric vehicle is difficult to predict, Monte Carlo simulation is used to predict the charging load curve of the electric vehicle; second, based on the various flexible resources inside the low-voltage flexible substation, an aggregated adjustable domain of the flexible low-voltage substation is constructed; finally, the aggregated adjustable domain is coordinated with the medium-voltage distribution network, and a second-order cone transformation is performed under the constraints of the active distribution network operation to minimize the line loss of the medium-voltage distribution network. The present invention not only does not involve the medium-voltage distribution network in the process of obtaining the aggregated adjustable domain of the interactive node, protecting the privacy of the low-voltage distribution network, but also can simplify the dispatching model of the medium-voltage distribution network and improve the dispatching efficiency.

Description

Technical Field

[0001] The present invention belongs to the technical field of coordinated dispatching of distribution networks, and in particular to a method for coordinated dispatching of medium and low voltage distribution networks based on flexible resources on the low voltage side. Background Art

[0002] Traditional distribution networks have fewer controllable devices and passively receive power from the main grid. Only some nodes contain a small amount of reactive compensation equipment (capacitor banks), and substation nodes may have on-load tap-changing transformers. With the access of large-scale distributed sources and loads (distributed photovoltaics, energy storage, electric vehicles, etc.), distribution networks are more proactive and can adjust their operating status according to the requirements of the main grid; sometimes they can reversely transmit power to the main grid and provide voltage support.

[0003] Low-voltage flexible substations refer to the configuration of distributed solar photovoltaic, distributed energy storage, temperature control loads, electric vehicles and other flexible resources in the interactive nodes of the low-voltage distribution network, and interact with the power system in a "load-follows-source" manner through flexible technologies on the energy user side. In addition, the aggregated adjustable domain of the low-voltage flexible substation is a set of operating working points of all state variables that meet the operating constraints and safety constraints of the flexible and controllable resources in the system. It is a high-dimensional complex space that is difficult to observe directly. Therefore, it is of great significance to solve the aggregated adjustable domain of the low-voltage substation. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a method for coordinated scheduling of medium and low voltage distribution networks based on flexible resources on the low voltage side.

[0005] The present invention is implemented by the following scheme:

[0006] A coordinated dispatching method for medium and low voltage distribution networks based on low voltage side flexible resources comprises the following steps:

[0007] Step 1: According to the charging load characteristics of electric vehicles, based on Monte Carlo simulation, predict the charging load curve of electric vehicles;

[0008] Step 2: Based on the various flexible resources inside the low-voltage flexible area, introduce equipment, state variables and decision variables, map the original high-dimensional operation domain to the two-dimensional state space of active power and reactive power at the exit of the low-voltage flexible area, and construct an aggregated adjustable domain of the flexible low-voltage area; the flexible resources include the electric vehicles in step 1;

[0009] Step 3: Based on the flexible low-voltage area aggregate adjustable domain obtained in step 2, active distribution network constraints are added, and the aggregate adjustable domain is coordinated with the medium-voltage distribution network with the minimum line loss of the medium-voltage distribution network as the optimization goal;

[0010] Step 4: Use the second-order cone change to correct the dynamic Distflow branch flow constraint, and use the expected value of the original objective function as the optimization target to achieve coordinated scheduling of medium and low voltage distribution networks.

[0011] During the coordinated dispatching of medium and low voltage distribution networks, the flexible resources within the low voltage substation are taken into consideration, a dispatching domain model is constructed, and the dispatchable domain of the interactive nodes is obtained. The whole process does not require the participation of the medium voltage distribution network, which can effectively protect the privacy of the low voltage distribution network. Secondly, the medium voltage distribution network is dispatched based on the dispatching domain given by the low voltage distribution network. There is no need to directly dispatch the numerous controllable devices under the low voltage distribution network, which can simplify the dispatching model of the medium voltage distribution network and improve the dispatching efficiency.

[0012] Based on the above description, the present invention first uses Monte Carlo simulation to predict the charging load curve of electric vehicles; secondly, based on the various flexible resources inside the low-voltage flexible area, an aggregated adjustable domain of the flexible low-voltage area is constructed; finally, the aggregated adjustable domain is coordinated with the medium-voltage distribution network, and under the constraints of the active distribution network operation, a second-order cone transformation is performed to minimize the line loss of the medium-voltage distribution network. The present invention not only effectively protects the privacy of the low-voltage distribution network, but also simplifies the scheduling model of the medium-voltage distribution network and improves the scheduling efficiency.

[0013] Further optimization, in step 1, the electric vehicle charging load characteristics are as follows:

[0014]

[0015]

[0016]

[0017]

[0018] P EV,t ≤P EV,max (A-5)

[0019]

[0020] Where: t sc The time when electric vehicle starts charging; μ sc and σ sc are the expected value and standard deviation of the probability density function of the starting charging time of the electric vehicle; x is the mileage of the electric vehicle on that day; μ sd and σ sd are the expected value and standard deviation of the probability density function of the daily mileage of electric vehicles; x max is the maximum mileage of an electric vehicle; T ch The charging time of electric vehicles; C soc C is the state of charge at the end of driving; soc,maxis the maximum state of charge of the electric vehicle; C EV is the battery capacity of the electric vehicle; EV is the charging efficiency; P EV,t is the charging power at time t; P EV,max is the upper limit of the active power of the charging pile; Q EV,t is the reactive power generated by the charging pile at time t.

[0021] Further optimization, predicting the charging load curve of the electric vehicle in step 1 specifically includes the following steps:

[0022] Step 1.1, initialize the number of electric vehicle batteries, maximum charging power, power consumption per kilometer and simulation times;

[0023] Step 1.2, randomly generate the starting charging time and daily mileage of a single electric vehicle, assume that the electric vehicle is charged to a full charge, calculate the charging time of the electric vehicle, and then obtain the end time of the electric vehicle charging;

[0024] Step 1.3, superimpose the charging load of each electric vehicle to obtain the charging load of all electric vehicles in a day;

[0025] Step 1.4, add up the charging load obtained in each simulation to get the total charging load, and calculate the average value of the electric vehicle charging load in each simulation as the charging load curve.

[0026] Further optimization, in step 2, constructing the low-voltage flexible stage area aggregation adjustable domain includes the following constraints:

[0027] 1) Energy storage system constraints: Formula (A-7) and Formula (A-8) are respectively the upper limit constraints of energy storage charging and discharging power; Formula (A-9) is the constraint of non-simultaneous charging and discharging; Formula (A-10) is the energy storage balance constraint; Formula (A-11) is the upper and lower limit constraints of energy storage;

[0028] 0≤P ch,t ≤β ch,t P ESS ,max (A-7)

[0029] 0≤P dis,t ≤β dis,t P ESS ,max (A-8)

[0030] β ch,t +β dis,t ≤1 (A-9)

[0031] E t =(1-δ)E t-1 +η ch P ch,t Δt-P dis,tΔt / η dis (A-10)

[0032] 0.1E N ≤E t ≤0.9E N (A-11)

[0033] Where: P ch,t and P dis,t are the charging and discharging power of energy storage during period t; P ESS,max is the maximum output and input power of energy storage; β ch,t and β dis,t are the energy storage charging and discharging states at time t, which are 1 when charging or discharging, and 0 otherwise; E t is the energy storage capacity at time t; η ch and η dis are the energy storage charging and discharging efficiency respectively; δ is the self-consumption rate of energy storage; Δt is the time interval between adjacent time periods; E N is the access rated capacity of the energy storage; the energy storage capacity at time t-1 of each scenario is set to be consistent.

[0034] 2) Photovoltaic system constraints: Formula (A-12) is the upper and lower limit constraints of photovoltaic output; Formula (A-13) calculates the amount of abandoned light at time t; Formula (A-14) is the upper limit constraint of the abandoned light rate; Formula (A-15) shows that the photovoltaic power generation system should meet the power factor that is dynamically adjustable within the range of leading 0.95 to lagging 0.95;

[0035] 0≤P PV,t ≤P PV,max,t (A-12)

[0036] P PV,cut,t =P PV,max,t -P PV,t (A-13)

[0037] P PV,cut,t ≤κ PV,max P PV,max,t (A-14)

[0038] -η PV,max P PV,t ≤Q PV,t ≤η PV,max P PV,t (A-15)

[0039] Where: P PV,max,t is the maximum output of photovoltaic power at time t; P PV,t is the actual output of photovoltaic power at time t; P PV,cut,t is the amount of abandoned light at time t; PV,max is the upper limit of the light abandonment rate; Q PV,tis the reactive power absorbed or emitted by the photovoltaic system at time t; η PV,max is a constant.

[0040] 3) HVAC operation constraints: Formula (A-16) is an upper limit constraint on the load power; Formula (A-17) is the load power factor constraint; Formula (A-18) ensures that the indoor temperature is within a range; Formula (A-19) represents a linear relationship between the indoor temperature at the next moment and the current indoor temperature, outdoor temperature and load power; Formula (A-20) is the HVAC inverter capacity constraint.

[0041] 0≤P HVAC,t ≤P HVAC,max (A-16)

[0042] Q HVAC,t =η HVAC ·P HVAC,t (A-17)

[0043] F in,min ≤F in,t ≤F in,max (A-18)

[0044] F in,t =F in,t-1 +α T ·(F out,t -F in,t-1 )+Δt·β T ·P HVAC,t (A-19)

[0045]

[0046] Where: P HVAC，t , Q HVAC，t is the active power and reactive power of the HVAC load at time t; P HVAC,max is the upper limit of HVAC active power; η HVAC is a constant, which specifies a fixed power factor; F in,t is the indoor temperature of the building at time t; F in,max 、F in,min are the highest and lowest temperatures of the room respectively; α T ∈(0,1) and β T is a thermal parameter that specifies the building and environmental characteristics; β T The value of includes positive and negative cases, where a positive value indicates heating and a negative value indicates cooling; S HVAC,max It is the maximum capacity of the HVAC inverter.

[0047] 4) AC / DC converter constraints: Formula (A-21) represents the efficiency conversion of power flowing through the VSC; Formula (A-22) represents the upper and lower limits of reactive power compensation constraints; Formula (A-23) represents the VSC capacity constraints;

[0048] P AC,t,VSC =P DC,t,VSC (A-21)

[0049] Q AC,t,VSC,min ≤Q AC,t,VSC ≤Q AC,t,VSC,max (A-22)

[0050]

[0051] Where: P AC，t，VSC , P DC，t，VSC are the active power values ​​of the AC and DC sides of the converter at time t; Q AC,t,VSC is the reactive power compensation value of the converter at time t; Q AC,t,VSC,max and Q AC,t,VSC,min Q AC,t,VSC The upper and lower limits of S AC,VSC is the capacity of the VSC.

[0052] 5) Energy conservation constraint: Equations (A-24) and (A-25) express the conservation of active power and reactive power respectively;

[0053]

[0054]

[0055] Where: P LVP,t and Q LVP,t They are active power and reactive power at the low voltage station gateway respectively; are the reactive power of the AC / DC converter connected to the energy storage device and the photovoltaic power generation device on the AC side; N EV is the total number of electric vehicles in the area; P h,EV,t represents the charging power of the hth electric vehicle at the charging station at time t; Q h,EV,t It represents the reactive power of the charging station of the h-th electric vehicle at time t.

[0056] 6) Synthesis of low-voltage flexible platform polymerization tunable domain:

[0057] Ω LVP ={(P LVP,t ,Q LVP,t )|h(W)=0,g(W)≤0} (A-26)

[0058] Where: W represents all operating state variables of the low voltage area, h(W) = 0 is the equality constraint; g(W) ≤ 0 is the inequality constraint; Ω LVP In order to meet the flexible and controllable resource operation constraints and safety constraints within the low-voltage substation, the aggregated adjustable domain of active power and reactive power at the low-voltage substation is established.

[0059] The upper and lower boundaries of reactive power compensation power at the low voltage station are obtained by using equations (A-27) and (A-28) respectively:

[0060]

[0061]

[0062] Where: and Q LPV,t are the upper and lower limits of reactive power compensation of the system at time t; Q LPV,t,c For the interval One of a series of working points obtained at equal distances within Fix reactive power compensation to Q LPV,t,c By repeatedly solving the following two equations, we can obtain a series of safety boundary points of the corresponding active power values ​​at the low-voltage station area.

[0063]

[0064]

[0065] Define C(Y) as the expression for the boundary points of the aggregate adjustable domain:

[0066]

[0067] Where: Q LPV,t,c For the interval One of a series of working points obtained at equal distances within the same region; Y is the set of boundary points of the aggregated adjustable domain; δ r is the rth segment coefficient; z r is the rth boundary working point; k z is the number of boundary points; z is the linear combination of boundary working points.

[0068] The active power and reactive power corresponding to the working point at the boundary of the aggregate adjustable domain can be expressed as follows:

[0069]

[0070]

[0071] Where: z r,P is the active power corresponding to the rth boundary operating point; is the linear combination of the active power at the boundary operating points; r,Q is the reactive power corresponding to the rth boundary operating point; is the linear combination of reactive power at the boundary operating points.

[0072] Further optimization, in step 3, the active distribution network constraints and objective functions are as follows:

[0073] 1) Distflow branch flow constraints:

[0074]

[0075]

[0076]

[0077]

[0078]

[0079] Where: set m(j) is the set of head nodes of the branch with j as the terminal node in the power grid; set n(j) is the set of terminal nodes of the branch with j as the head node in the power grid; set u(j) is the non-station node; set v(j) is the station node; U j,t is the voltage amplitude of node j at time t; P ij,t and Q ij,t are respectively the active power and reactive power at the head end of branch ij; P j,t and Q j,t are the net active power and net reactive power input values ​​of node j respectively; P LVP,j,t and Q LVP,j,t are respectively the active power and reactive power at the low voltage station gateway at node j; P load,j,t and Q load,j,t are load active power and load reactive power respectively; r ij and x ij are the branch resistance and reactance respectively.

[0080] 2) System safe operation constraints:

[0081] U j,min ≤U j,t ≤U j,max (A-39)

[0082] I ij,t ≤I ij ,max (A-40)

[0083] Where: U j,max and U j,minare the upper and lower limits of the voltage at node j respectively; I ij,t is the upper limit of the current of line ij.

[0084] 3) Root node injection power constraint:

[0085] P root,min ≤P root,t ≤P root,max (A-41)

[0086] Q root,min ≤Q root,t ≤Q root,max (A-42)

[0087] Where: P root,t and Q root,t are the active power and reactive power flowing from the root node into the low-voltage distribution network respectively; P root,max and P root,min are the upper and lower limits of the active power flowing from the root node into the low-voltage distribution network; Q root,max and Q root,min They are the upper and lower limits of the reactive power flowing from the root node into the low-voltage distribution network.

[0088] 4) Main function:

[0089]

[0090] Where: Δt is the time interval; N is the total number of nodes in the power grid; set n(i) is the set of end nodes of the branch with i as the head node in the power grid; P loss is the total line loss.

[0091] Further optimization, in step 4, the second-order cone transformation is as follows:

[0092] Definition (A-44):

[0093]

[0094] Formula (A-34), formula (A-35), formula (A-38), formula (A-44), and formula (A-43) are rewritten as follows:

[0095]

[0096]

[0097]

[0098]

[0099]

[0100] Where:

[0101] Compared with the prior art, the present invention has the following beneficial effects:

[0102] The coordinated dispatching method of medium and low voltage distribution networks based on flexible resources on the low voltage side proposed in the present invention mainly considers the following two aspects: one is the construction of the low voltage flexible substation aggregation adjustable domain, and the other is taking into account the constraints of the active distribution network to achieve coordinated dispatching of the medium and low voltage distribution network systems to minimize line losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0103] Figure 1 is a flow chart of the method of the present invention;

[0104] Figure 2 This is a diagram of a calculation example of hierarchical coordinated dispatching of medium and low voltage distribution networks;

[0105] Figure 3 It is the aggregate adjustable domain of the three low-voltage stations under normal conditions and their current operating points;

[0106] Figure 4 It is the aggregate adjustable domain of the three low-voltage stations and their current operating points when the photovoltaic output is low;

[0107] Figure 5 It is the aggregate adjustable domain of the three low-voltage sections and their current operating points when the energy storage is only charged. DETAILED DESCRIPTION

[0108] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention. After reading the present invention, modifications of various equivalent forms of the present invention by those skilled in the art all fall within the scope defined by the appended claims of the present application.

[0109] like Figure 1 As shown, a method for coordinated scheduling of medium and low voltage distribution networks based on low voltage side flexible resources comprises the following steps:

[0110] Step 1: According to the charging load characteristics of electric vehicles, the charging load curve of electric vehicles is predicted based on Monte Carlo simulation.

[0111] The electric vehicle charging load characteristics are as follows:

[0112]

[0113]

[0114]

[0115]

[0116] P EV,t ≤P EV,max (B-5)

[0117]

[0118] Where: t sc The time when electric vehicle starts charging; μ sc and σ sc are the expected value and standard deviation of the probability density function of the starting charging time of the electric vehicle; x is the mileage of the electric vehicle on that day; μ sd and σ sd are the expected value and standard deviation of the probability density function of the daily mileage of electric vehicles; x max is the maximum mileage of an electric vehicle; T ch The charging time of electric vehicles; C soc C is the state of charge at the end of driving; soc,max is the maximum state of charge of the electric vehicle; C EV is the battery capacity of the electric vehicle; EV is the charging efficiency; P EV,t is the charging power at time t; P EV,max is the upper limit of the active power of the charging pile; Q EV,t is the reactive power generated by the charging pile at time t.

[0119] Formula (B-1) is the probability density function of the starting charging time of electric vehicles fitted based on the relevant data of automobile travel in the National Household Travel Survey (NHTS); Formula (B-2) is the probability density function of the daily mileage of electric vehicles; Formula (B-3) calculates the state of charge of the battery when the driving ends and charging begins; Formula (B-4) calculates the charging time of the electric vehicle; Formula (B-5) is the upper limit constraint of the active power of the charging pile; Formula (B-6) represents the capacity constraint.

[0120] The specific steps for simulating and fitting the electric vehicle charging load curve are as follows:

[0121] Step 1.1, initialize the number of electric vehicle batteries, maximum charging power, power consumption per kilometer and simulation times;

[0122] Step 1.2, randomly generate the starting charging time and daily mileage of a single electric vehicle, assume that the electric vehicle is charged to a full charge, calculate the charging time of the electric vehicle, and then obtain the end time of the electric vehicle charging;

[0123] Step 1.3, superimpose the charging load of each electric vehicle to obtain the charging load of all electric vehicles in a day;

[0124] Step 1.4, add up the charging load obtained in each simulation to get the total charging load, and calculate the average value of the electric vehicle charging load in each simulation as the charging load curve.

[0125] Step 2: Based on the various flexible resources inside the low-voltage flexible area, introduce equipment, state variables and decision variables, map the original high-dimensional operation domain to the two-dimensional state space of active power and reactive power at the exit of the low-voltage flexible area, and construct an aggregated adjustable domain of the flexible low-voltage area; the flexible resources include the electric vehicles in step 1. The following constraints are included:

[0126] 1) Energy storage system constraints

[0127] 0≤P ch,t ≤β ch,t P ESS,max (B-7)

[0128] 0≤P dis,t ≤β dis,t P ESS,max (B-8)

[0129] β ch,t +β dis,t ≤1 (B-9)

[0130] E t =(1-δ)E t-1 +η ch P ch,t Δt-P dis,t Δt / η dis (B-10)

[0131] 0.1E N ≤E t ≤0.9E N (B-11)

[0132] Where: P ch,t and P dis,t are the charging and discharging power of the energy storage during period t; P ESS,max is the maximum output and input power of energy storage; β ch,t and β dis,t are the energy storage charging and discharging states at time t, which are 1 when charging or discharging, and 0 otherwise; E t is the energy storage capacity at time t; η ch and η dis are the energy storage charging and discharging efficiency respectively; δ is the self-consumption rate of energy storage; Δt is the time interval between adjacent time periods; E N is the access rated capacity of the energy storage; the energy storage capacity at time t-1 of each scenario is set to be consistent.

[0133] Formula (B-7) and Formula (B-8) are respectively the upper limit constraints of energy storage charging and discharging power; Formula (B-9) is the constraint of energy storage charging and discharging at different times; Formula (B-10) is the energy storage balance constraint; Formula (B-11) is the upper and lower limit constraints of energy storage.

[0134] 2) Photovoltaic system constraints

[0135] 0≤P PV,t ≤P PV,max,t (B-12)

[0136] P PV,cut,t =P PV,max,t -P PV,t (B-13)

[0137] P PV,cut,t ≤κ PV,max P PV,max,t (B-14)

[0138] -η PV,max P PV,t ≤Q PV,t ≤η PV,max P PV,t (B-15)

[0139] Where: P PV,max,t is the maximum output of photovoltaic power at time t; P PV,t is the actual output of photovoltaic power at time t; P PV,cut,t is the amount of abandoned light at time t; PV,max is the upper limit of the light abandonment rate; Q PV,t is the reactive power absorbed or emitted by the photovoltaic system at time t; η PV,max is a constant.

[0140] Formula (B-12) is the upper and lower limit constraints of photovoltaic output; Formula (B-13) calculates the amount of abandoned light at time t; Formula (B-14) is the upper limit constraint of the abandoned light rate; Formula (B-15) shows that the photovoltaic power generation system should satisfy the power factor that is dynamically adjustable in the range of leading 0.95 to lagging 0.95.

[0141] 3) Heating, Ventilation and Air Conditioning (HVAC) Operation Constraints

[0142] 0≤P HVAC,t ≤P HVAC,max (B-16)

[0143] Q HVAC,t =η HVAC ·P HVAC,t (B-17)

[0144] F in,min ≤F in,t ≤F in,max (B-18)

[0145] F in,t =F in,t-1 +α T ·(F out,t -F in,t-1 )+Δt·β T ·P HVAC,t (B-19)

[0146]

[0147] Where: P HVAC,t , Q HVAC,t is the active power and reactive power of the HVAC load at time t; P HVAC,max is the upper limit of HVAC active power; η HVAC is a constant, which specifies a fixed power factor; F in,t is the indoor temperature of the building at time t; F in,max 、F in,min are the highest and lowest temperatures of the room respectively; α T ∈(0,1) and β T is a thermal parameter that specifies the building and environmental characteristics; β T Can be positive or negative, positive for heating (winter) and negative for cooling (summer). HVAC,max It is the maximum capacity of the HVAC inverter.

[0148] Formula (B-16) is an upper limit constraint on the load power; Formula (B-17) is the power factor constraint of the load; Formula (B-18) ensures that the indoor temperature is within a range; Formula (B-19) represents a linear relationship between the indoor temperature at the next moment and the current indoor temperature, outdoor temperature and load power; Formula (B-20) is the HVAC inverter capacity constraint.

[0149] 4) AC / DC converter constraints

[0150] P AC,t,VSC =P DC,t,VSC (B-21)

[0151] Q AC,t,VSC,min ≤Q AC,t,VSC ≤Q AC,t,VSC,max (B-22)

[0152]

[0153] Where: P AC,t,VSC and P DC,t,VSC are the active power values ​​of the AC and DC sides of the converter at time t; Q AC,t,VSC is the reactive power compensation value of the converter at time t; Q AC,t,VSC,max and QAC,t,VSC,min Q AC,t,VSC The upper and lower limits of S AC,VSC is the capacity of the VSC.

[0154] Formula (B-21) represents the efficiency conversion of power flowing through the VSC; Formula (B-22) represents the upper and lower limit constraints of reactive power compensation; Formula (B-23) represents the VSC capacity constraint.

[0155] 5) Energy conservation constraints

[0156]

[0157]

[0158] Where: P LVP，t , Q LVP，t They are active power and reactive power at the low voltage station gateway respectively; are the reactive power of the AC / DC converter connected to the energy storage device and the photovoltaic power generation device on the AC side; N EV is the total number of electric vehicles in the area; P h,EV,t It represents the charging power of the hth electric vehicle at the charging station at time t. h,EV,t It represents the reactive power of the charging station of the h-th electric vehicle at time t.

[0159] Equations (B-24) and (B-25) respectively express the conservation of active power and reactive power.

[0160] 6) Synthesis of the aggregated adjustable domain of the low-voltage flexible area. W represents all the operating state variables of the low-voltage area. The aggregated adjustable domain of the low-voltage area is expressed as follows:

[0161] Ω LVP ={(P LVP,t ,Q LVP,t )|h(W)=0,g(W)≤0} (B-26)

[0162] Where: h(W) = 0 is the equality constraint; g(W) ≤ 0 is the inequality constraint; Ω LVP In order to meet the flexible and controllable resource operation constraints and safety constraints within the low-voltage substation, the aggregated adjustable domain of active power and reactive power at the low-voltage substation is established.

[0163] The upper and lower boundaries of reactive power compensation power at the low voltage station are obtained by using equations (B-27) and (B-28) respectively:

[0164]

[0165]

[0166] Where: and Q LPV,t They are the upper and lower boundaries of the system reactive power compensation at time t respectively.

[0167] Q LPV,t,c For the interval One of a series of working points obtained at equal distances within Fix reactive power compensation to Q LPV,t,c Repeatedly solving equations (39-40) can obtain a series of safety boundary points of active power values ​​at the corresponding low-voltage station area:

[0168]

[0169]

[0170] Based on the obtained series of working points, an aggregate adjustable domain based on the convex envelope can be constructed. The set of boundary points of the aggregate adjustable domain is defined as Y, and the expression of the boundary points of the aggregate adjustable domain C(Y) is:

[0171]

[0172] Where: r is the rth segment coefficient; z r is the rth boundary working point; k z is the number of boundary points; z is the linear combination of boundary working points.

[0173] The active power and reactive power corresponding to the working point at the boundary of the aggregate adjustable domain can be written as Equation (B-32) and Equation (B-33):

[0174]

[0175]

[0176] Where: z r,P is the active power corresponding to the rth boundary operating point; is the linear combination of the active power at the boundary operating points; r,Q is the reactive power corresponding to the rth boundary operating point; is a linear combination of the reactive power at the boundary operating points.

[0177] Aggregating the adjustable domain is essentially a process of linearizing the nonlinear safety boundary piecewise. Based on a series of safety domain boundary points, the flexible operation domain can be generated with the help of MATLAB's ConvexHull function.

[0178] Step 3: Based on the low-voltage flexible area aggregate adjustable domain obtained in step 2, add active distribution network constraints, take the minimum medium-voltage distribution network line loss as the optimization goal, and coordinate the aggregate adjustable domain with the medium-voltage distribution network. The specific method is as follows:

[0179] 1) Distflow branch flow constraints

[0180]

[0181]

[0182]

[0183]

[0184]

[0185] Where: set m(j) is the set of head nodes of the branch with j as the terminal node in the power grid; set n(j) is the set of terminal nodes of the branch with j as the head node in the power grid; set u(j) is the non-station node; set v(j) is the station node; U j,t is the voltage amplitude of node j at time t; P ij,t and Q ij,t are respectively the active power and reactive power at the head end of branch ij; P j,t and Q j,t are the net active power and net reactive power input values ​​of node j respectively; P LVP,j,t and Q LVP,j,t are respectively the active power and reactive power at the low voltage station gateway at node j; P load,j,t and Q load,j,t are load active power and load reactive power respectively; r ij and x ij are the branch resistance and reactance respectively.

[0186] 2) System safety operation constraints

[0187] U j,min ≤U j,t ≤U j,max (B-39)

[0188] I ij,t ≤I ij,max (B-40)

[0189] Where: U j,max and U j,min are the upper and lower limits of the voltage at node j respectively; I ij,t is the upper limit of the current of line ij.

[0190] Formula (B-39) is the upper and lower limit constraints of the node voltage; Formula (B-40) is the line current constraint.

[0191] 3) Root node injection power constraint

[0192] P root,min ≤P root,t ≤P root,max (B-41)

[0193] Q root,min ≤Q root,t ≤Q root,max (B-42)

[0194] Where: P root,t and Q root,t are the active power and reactive power flowing from the root node into the low-voltage distribution network respectively; P root,max and P root,min are the upper and lower limits of the active power flowing from the root node into the low-voltage distribution network; Q root,max and Q root,min They are the upper and lower limits of the reactive power flowing from the root node into the low-voltage distribution network.

[0195] Formula (B-41) is the upper and lower limit constraints of the active power injected into the node; Formula (B-42) is the upper and lower limit constraints of the reactive power injected into the node.

[0196] 4) Main function

[0197]

[0198] Where: Δt is the time interval; N is the total number of nodes in the power grid; set n(i) is the set of end nodes of the branch with i as the head node in the power grid; P loss is the total line loss.

[0199] Step 4: Use the second-order cone change to correct the dynamic Distflow branch flow constraint, and use the expected value of the original objective function as the optimization target to achieve coordinated scheduling of medium and low voltage distribution networks. The specific method is as follows:

[0200] Definition (B-44):

[0201]

[0202] make Formula (B-34), formula (B-35), formula (B-38), and formula (B-44) are further rewritten as formula (B-45)-(B-48):

[0203]

[0204]

[0205]

[0206]

[0207] Formula (B-45)-(B-47) is a system of linear equations. Formula (B-48) is a quadratic equation in a concise form. The main function formula (B-43) can be rewritten as:

[0208]

[0209] The present invention proposes a medium- and low-voltage distribution network coordinated scheduling model based on low-voltage side flexible resources, taking formula (B-43) as the optimization target, satisfying electric vehicle charging load characteristic constraints (B-1)-(B-6), energy storage system constraints (B-7)-(B-11), photovoltaic system constraints (B-12)-(B-15), air conditioning operation constraints (B-16)-(B-20), AC / DC converter constraints (B-21)-(B-23), energy conservation constraints (B- -24)-(B-25), low-voltage flexible substation aggregate adjustable domain synthesis constraints (B-26)-(B-33), Distflow branch flow constraints (B-34)-(B-38), system safety operation constraints (B-39)-(B-40), root node injection power constraints (B-41)-(B-42), Distflow branch flow constraints and optimization objectives after second-order cone transformation (B-45)-(B-48), (B-49).

[0210] Case analysis:

[0211] like Figure 2 As shown, this example selects the IEEE33-node radial distribution system, which contains 32 branches and does not consider 5 permanent branches. The voltage level of the medium voltage distribution network is 10kV, and the voltage level of the low voltage distribution network is 380V. Flexible low-voltage substations are connected at nodes 8, 22, and 32. The internal configuration of each flexible low-voltage substation is shown in Table 1, where " / " represents no configuration. The time of 15:00 is selected for analysis. At this time, the outdoor temperatures of substations 1 to 3 are 14°C, 13°C, and 15°C, respectively. The initial indoor temperature is 23°C. The root node voltage is fixed at 1. The initial SOC of the battery energy storage is set to 0.5. The present invention is implemented through the GAMS optimization platform, and the GUROBI solver is used to solve the QCP problem.

[0212] Table 1 Internal configuration of three flexible low-voltage areas

[0213]

[0214] Figure 3The aggregate adjustable domain of each flexible substation is shown, and the five-pointed star mark is the working point of each flexible substation at the optimal solution. Substation 1 works at the boundary of the aggregate adjustable domain, while substations 2 and 3 work inside the aggregate adjustable domain. In the figure, positive data indicates absorption from the power grid (active power or reactive power), and negative data indicates emission to the power grid (active power or reactive power). Since the photovoltaic output in the low-voltage substation is large and the energy storage charge state is high, the three substations all provide a certain amount of active power to the medium-voltage distribution network to achieve on-site balance of the active power of the medium-voltage distribution network and reduce the node voltage offset. The other three substations all emit reactive power to the power grid to achieve reactive compensation. This is used as the control group in the calculation example.

[0215] Figure 4 It shows that under the influence of natural factors, the photovoltaic output will be relatively small, and the aggregate adjustable range and current working point of each flexible station area will be relatively small. Compared with the control group, Figure 4 The lower and upper limits of the active power of the aggregated adjustable domain of each flexible substation are both increased, that is, the maximum active power that can be sent to the grid is reduced, while the maximum active power that needs to be absorbed from the grid is increased; at the same time, the lower limit of the reactive power of the aggregated adjustable domain of each flexible substation is increased, and the upper limit is reduced, that is, the maximum reactive power that can be sent to the grid and the maximum reactive power that needs to be absorbed from the grid are both reduced. Overall, the aggregated adjustable domain of the flexible substation "shrinks" and "shifts" to the right. From the perspective of dispatching strategy, due to the small photovoltaic output in the low-voltage substation, each substation absorbs as little active power as possible from the medium-voltage distribution network. At the same time, each substation sends out as much reactive power as possible to maintain voltage stability.

[0216] Figure 5 The aggregated adjustable domain and current working point of each flexible substation when energy storage is only allowed to charge are shown. The lower limit of active power of substations 2 and 3 becomes larger, that is, the maximum active power that can be sent to the grid is reduced. Overall, the left end of the aggregated adjustable domain of substations 2 and 3 shows a "shrinkage" phenomenon compared with the control group; the right end of the aggregated adjustable domain of substations 2 and 3 is basically unchanged compared with the control group. From the perspective of scheduling strategy, when only energy storage charging is allowed, since the maximum active power that substations 2 and 3 can send to the grid is reduced, their working points are located at the left boundary of the aggregated adjustable domain, that is, substations 2 and 3 are required to provide active power as much as possible.

[0217] In summary, the data of the flexible station working points in each scenario in the example are shown in Table 2:

[0218] Table 2 Data of flexible station working points in each scenario

[0219]

[0220] The above description is only a specific implementation of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a technician familiar with the technical field within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention.

Claims

1. A coordinated dispatching method for medium and low voltage distribution networks based on low voltage side flexible resources, characterized in that: The following steps are involved: Step 1: According to the charging load characteristics of electric vehicles, based on Monte Carlo simulation, predict the charging load curve of electric vehicles; Step 2: Based on the various flexible resources inside the low-voltage flexible area, introduce equipment, state variables and decision variables, map the original high-dimensional operation domain to the two-dimensional state space of active power and reactive power at the exit of the low-voltage flexible area, and construct an aggregated adjustable domain of the flexible low-voltage area; the flexible resources include the electric vehicles in step 1; Step 3: Based on the flexible low-voltage area aggregate adjustable domain obtained in step 2, active distribution network constraints are added, and the aggregate adjustable domain is coordinated with the medium-voltage distribution network with the minimum line loss of the medium-voltage distribution network as the optimization goal; Step 4: Use the second-order cone change to correct the dynamic Distflow branch power flow constraint, and use the expected value of the original objective function as the optimization target to achieve coordinated scheduling of medium and low voltage distribution networks; In step 1, the electric vehicle charging load characteristics are as follows: P EV,t ≤P EV,max (A-5) Where: t sc The time when electric vehicle starts charging; μ sc and σ sc are the expected value and standard deviation of the probability density function of the starting charging time of the electric vehicle; x is the mileage of the electric vehicle on that day; μ sd and σ sd are the expected value and standard deviation of the probability density function of the daily mileage of electric vehicles; x max is the maximum mileage of an electric vehicle; T ch The charging time of electric vehicles; C soc C is the state of charge at the end of driving; soc,max is the maximum state of charge of the electric vehicle; C EV is the battery capacity of the electric vehicle; EV is the charging efficiency; P EV,t is the charging power at time t; P EV,max is the upper limit of the active power of the charging pile; Q EV,t is the reactive power generated by the charging pile at time t; The step 1 of predicting the charging load curve of the electric vehicle specifically includes the following steps: Step 1.1, initialize the number of electric vehicle batteries, maximum charging power, power consumption per kilometer and simulation times; Step 1.2, randomly generate the starting charging time and daily mileage of a single electric vehicle, assume that the electric vehicle is charged to a full charge, calculate the charging time of the electric vehicle, and then obtain the end time of the electric vehicle charging; Step 1.3, superimpose the charging load of each electric vehicle to obtain the charging load of all electric vehicles in a day; Step 1.4, add up the charging load obtained in each simulation to obtain the total charging load, and calculate the average value of the electric vehicle charging load in each simulation as the charging load curve; In step 2, constructing a low-voltage flexible stage area aggregation adjustable domain includes the following constraints: 1) Energy storage system constraints: Formula (A-7) and Formula (A-8) are respectively the upper limit constraints of energy storage charging and discharging power; Formula (A-9) is the constraint of non-simultaneous charging and discharging; Formula (A-10) is the energy storage balance constraint; Formula (A-11) is the upper and lower limit constraints of energy storage; 0≤P ch,t ≤β ch,t P ESS,max (A-7) 0≤P dis,t ≤β dis,t P ESS,max (A-8) β ch,t +β dis,t ≤1 (A-9) E t =(1-δ)E t-1 +n ch P ch,t Δt-P dis,t Δt / h dis (A-10) 0.1E N ≤E t ≤0.9E N (A-11) Where: P ch,t and P dis,t are the charging and discharging power of the energy storage during period t; P ESS,max is the maximum output and input power of energy storage; β ch,t Indicates the energy storage charging state at time t, which is 1 when charging and 0 otherwise; β dis,t Indicates the energy storage discharge state at time t, which is 1 when discharging and 0 otherwise; E t is the energy storage capacity at time t; η ch and η dis are the energy storage charging and discharging efficiency respectively; δ is the self-consumption rate of energy storage; Δt is the time interval between adjacent time periods; E N is the access rated capacity of the energy storage; the energy storage capacity at time t-1 of each scenario is set to be consistent; 2) Photovoltaic system constraints: Formula (A-12) is the upper and lower limit constraints of photovoltaic output; Formula (A-13) calculates the amount of abandoned light at time t; Formula (A-14) is the upper limit constraint of the abandoned light rate; Formula (A-15) shows that the photovoltaic power generation system should meet the power factor that is dynamically adjustable within the range of leading 0.95 to lagging 0.95; 0≤P PV,t ≤P PV,max,t (A-12) P PV,cut,t =P PV,max,t -P PV,t (A-13) P PV,cut,t ≤κ PV,max P PV,max,t (A-14) -η PV,max P.S PV,t ≤Q PV,t ≤η PV,max P.S PV,t (A-15) Where: P PV,max,t is the maximum output of photovoltaic power at time t; P PV,t is the actual output of photovoltaic power at time t; P PV,cut,t is the amount of abandoned light at time t; κ PV,max is the upper limit of the light abandonment rate; Q PV,t is the reactive power absorbed or emitted by the photovoltaic system at time t; η PV,max is a constant; 3) HVAC operation constraints: Formula (A-16) is an upper limit constraint on load power; Formula (A-17) is a load power factor constraint; Formula (A-18) ensures that the indoor temperature is within a certain range; Formula (A-19) represents a linear relationship between the indoor temperature at the next moment and the current indoor temperature, outdoor temperature and load power; Formula (A-20) is a HVAC inverter capacity constraint; 0≤P HVAC,t ≤P HVAC,max (A-16) Q HVAC,t =the HVAC ·P HVAC,t (A-17) F in,min ≤F in,t ≤F in,max (A-18) F in,t =F in,t-1 +α T ·(F out,t -F in,t-1 )+Δt·β T ·P HVAC,t (A-19) Where: P HVAC,t , Q HVAC,t is the active power and reactive power of the HVAC load at time t; P HVAC,max is the upper limit of active power of HVAC; η HVAC is a constant, that is, it stipulates a fixed power factor; F in,t is the indoor temperature of the building at time t; F in,max 、F in,min are the highest and lowest temperatures of the room respectively; α T ∈(0,1) and β T is a thermal parameter that specifies the building and environmental characteristics; β T The value of includes positive and negative cases, where a positive value indicates heating and a negative value indicates cooling; S HVAC,max is the maximum capacity of the HVAC inverter; 4) AC / DC converter constraints: Formula (A-21) represents the efficiency conversion of power flowing through the VSC; Formula (A-22) represents the upper and lower limits of reactive power compensation constraints; Formula (A-23) represents the VSC capacity constraints; P AC,t,VSC =P DC,t,VSC (A-21) Q AC,t,VSC,min ≤Q AC,t,VSC ≤Q AC,t,VSC,max (A-22) Where: P AC,t,VSC and P DC,t,VSC are the active power values ​​of the AC and DC sides of the converter at time t; Q AC,t,VSC is the reactive power compensation value of the converter at time t; Q AC,t,VSC,max and Q AC,t,VSC,min Q AC,t,VSC The upper and lower limits of S AC,VSC is the capacity of the VSC; 5) Energy conservation constraint: Equations (A-24) and (A-25) express the conservation of active power and reactive power respectively; Where: P LVP,t and Q LVP,t They are active power and reactive power at the low voltage station gateway respectively; are the reactive power of the AC / DC converter connected to the energy storage device and the photovoltaic power generation device on the AC side; N EV is the total number of electric vehicles in the area; P h,EV,t represents the charging power of the hth electric vehicle at the charging station at time t; Q h,EV,t represents the reactive power of the charging station of the hth electric vehicle at time t; 6) Synthesis of low-voltage flexible platform polymerization tunable domain: Ω LVP ={(P LVP,t ,Q LVP,t )|h(W)=0,g(W)≤0} (A-26) Where: W represents all operating state variables of the low voltage area, h(W) = 0 is the equality constraint; g(W) ≤ 0 is the inequality constraint; Ω LVP To meet the flexible and controllable resource operation constraints and safety constraints in the low-voltage area, the aggregate adjustable domain of active power and reactive power at the low-voltage area gateway; The upper and lower boundaries of reactive power compensation power at the low voltage station are obtained by using equations (A-27) and (A-28) respectively: Where: and Q LPV,t are the upper and lower limits of reactive power compensation of the system at time t; Q LPV,t,c For the interval One of a series of working points obtained at equal distances within Fix reactive power compensation to Q LPV,t,c Repeatedly solving the following two equations can obtain a series of safety boundary points of the active power value at the corresponding low-voltage station area: Define C(Y) as the expression for the boundary points of the aggregate adjustable domain: Where: Q LPV，t，c For the interval One of a series of working points obtained at equal distances within the same region; Y is the set of boundary points of the aggregated adjustable domain; δ r is the rth segment coefficient; z r is the rth boundary working point; k z is the number of boundary points; z is the linear combination of boundary working points; The active power and reactive power corresponding to the working point at the boundary of the aggregate adjustable domain are expressed as follows: Where: z r,P is the active power corresponding to the rth boundary operating point; is the linear combination of the active power at the boundary operating points; r,Q is the reactive power corresponding to the rth boundary operating point; is the linear combination of reactive power at boundary operating points; In step 3, the active distribution network constraints and objective functions are as follows: 1) Distflow branch flow constraints: Where: set m(j) is the set of head nodes of the branch with j as the terminal node in the power grid; set n(j) is the set of terminal nodes of the branch with j as the head node in the power grid; set u(j) is the non-station node; set v(j) is the station node; U j,t is the voltage amplitude of node j at time t; P ij,t and Q ij,t are respectively the active power and reactive power at the head end of branch ij; P j,t and Q j,t are the net active power and net reactive power input values ​​of node j respectively; P LVP,j,t and Q LVP,j,t are respectively the active power and reactive power at the low voltage station gateway at node j; P load,j,t and Q load,j,t are load active power and load reactive power respectively; r ij and x ij are branch resistance and reactance respectively; 2) System safe operation constraints: U j,min ≤U j,t ≤U j,max (A-39) I ij,t ≤I ij,max (A-40) Where: U j,max and U j,min are the upper and lower limits of the voltage at node j respectively; I ij,t is the upper limit of the current of line ij; 3) Root node injection power constraint: P root,min ≤P root,t ≤P root,max (A-41) Q root,min ≤Q root,t ≤Q root,max (A-42) Where: P root,t and Q root,t are the active power and reactive power flowing from the root node into the low-voltage distribution network respectively; P root,max and P root,min are the upper and lower limits of the active power flowing from the root node into the low-voltage distribution network; Q root,max and Q root,min They are the upper and lower limits of the reactive power flowing from the root node into the low-voltage distribution network, respectively; 4) Main function: Where: Δt is the time interval; N is the total number of nodes in the power grid; set n(i) is the set of end nodes of the branch with i as the head node in the power grid; P loss is the total line loss; In step 4, the second-order cone is transformed as follows: Definition (A-44): Formula (A-34), formula (A-35), formula (A-38), formula (A-44), and formula (A-43) are rewritten as follows: Where:

Images