An optimization scheduling method and terminal of a distributed smart grid

Abstract

The application discloses a kind of distributed intelligent power grid optimization scheduling method and terminal, for each independent dispatching area of distributed intelligent power grid, with the minimum operating cost of each independent dispatching area as target, establish the local power distribution network optimization scheduling model corresponding to each independent dispatching area, then with the voltage of adjacent independent dispatching area tie line two sides as consistency variable, using local power distribution network optimization scheduling model constructs global distributed intelligent power grid optimization scheduling model, using alternating direction multiplier method to solve standard alternating iteration problem, to obtain the scheduling strategy of distributed intelligent power grid. Therefore each independent region only and adjacent region carry out the interaction of voltage information, can well meet the privacy of source load information and operating information in each independent operation subject, and consider distributed power supply, switchable capacitor, on-load voltage regulating transformer tap and other intelligent power grid regulation measures, can better realize distributed optimization scheduling.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent power grid optimal scheduling, in particular to a distributed intelligent power grid optimal scheduling method and terminal. BACKGROUND

[0002] With the continuous development of new power systems, the form of future distribution networks will be distributed intelligent power grids. Distributed intelligent power grids have two important features:

[0003] First, under the reform of the power system, the operation subjects of the distribution network are highly diversified, such as incremental distribution networks, microgrids, virtual power plants, etc. Each independent subject has certain requirements for information privacy, and the internal dispatching center will make scheduling behavior in line with its own interests. The traditional centralized scheduling mode of the distribution network by the dispatching department of the power grid company will no longer be sustainable, and the distribution network scheduling mode will change from centralized to distributed.

[0004] Second, with the gradual popularization of secondary equipment coverage, more source-grid-load-storage resources will participate in regulation, further increasing the carrying capacity of the power grid.

[0005] Therefore, it is of great practical significance to study the optimal scheduling model and solution method of distributed intelligent power grids. SUMMARY

[0006] The technical problem to be solved by the present application is to provide a distributed intelligent power grid optimal scheduling method and terminal, which can improve the rationality of intelligent power grid distributed scheduling and the data privacy of distributed subjects.

[0007] To solve the above technical problems, the technical scheme adopted by the present application is:

[0008] A distributed intelligent power grid optimal scheduling method, comprising the steps of:

[0009] For each independent dispatching area of the distributed intelligent power grid, a local distribution network optimal scheduling model corresponding to each independent dispatching area is established with the minimum operation cost of each independent dispatching area as the target. The decision variables of the local distribution network optimal scheduling model include the output of the local distributed power source, the output of the switchable capacitor, and the position of the on-load voltage regulating transformer tap;

[0010] The voltages on both sides of the tie line of adjacent independent dispatching areas are used as consistency variables, and the local distribution network optimal scheduling model is used to construct a global distributed intelligent power grid optimal scheduling model;

[0011] The global distributed intelligent power grid optimal scheduling model is converted into a standard alternating iteration problem;

[0012] The standard alternating iteration problem is solved by using an alternating direction multiplier method to obtain a scheduling strategy of the distributed smart grid.

[0013] To solve the above technical problems, the application adopts another technical solution:

[0014] An optimization scheduling terminal of a distributed smart grid comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements each step of the optimization scheduling method of the distributed smart grid.

[0015] The application has the beneficial effect that for each independent scheduling area of the distributed smart grid, a local power distribution network optimization scheduling model corresponding to each independent scheduling area is established with the minimum operation cost of each independent scheduling area as the target, then the voltage on both sides of the tie line of adjacent independent scheduling areas is taken as a consistency variable, the local power distribution network optimization scheduling model is used to build a global distributed smart grid optimization scheduling model, the standard alternating iteration problem is solved by using an alternating direction multiplier method to obtain a scheduling strategy of the distributed smart grid. Therefore, each independent area only interacts with adjacent areas in voltage information, and the decoupling of each independent operation subject is realized through the tie line, which can well meet the privacy of the source and load information and operation information of each independent operation subject. In addition, the distributed power supply, switchable capacitor, on-load voltage regulating transformer tap and other smart grid regulation measures are considered in the application, which can improve the potential of the power grid and realize distributed optimization scheduling. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 A flowchart of an optimization scheduling method of a distributed smart grid according to an embodiment of the application;

[0017] Figure 2 A schematic diagram of an optimization scheduling terminal of a distributed smart grid according to an embodiment of the application;

[0018] Figure 3 A system topology diagram of a 3-partition IEEE 33-node power distribution network used in Embodiment 2 of the application;

[0019] Figure 4 A typical daily residential load curve diagram used in Embodiment 2 of the application;

[0020] Figure 5 A typical daily commercial load curve diagram used in Embodiment 2 of the application;

[0021] Figure 6 A typical daily industrial load curve diagram used in Embodiment 2 of the application;

[0022] Figure 7A typical daily wind power output curve diagram used in the second embodiment of the present application;

[0023] Figure 8 A typical daily photovoltaic power output curve diagram used in the second embodiment of the present application;

[0024] Figure 9 A diagram showing the active power output of various power sources in the second embodiment of the present application;

[0025] Figure 10 A diagram showing the reactive power output of various power sources in the second embodiment of the present application;

[0026] Figure 11 A diagram showing the operation of switchable capacitors in the second embodiment of the present application;

[0027] Label explanation:

[0028] 1. An optimal dispatch terminal of a distributed smart grid; 2. a memory; 3. a processor. DETAILED DESCRIPTION

[0029] To explain the technical content, the achieved purposes and effects of the present application in detail, the following will be described in combination with the embodiments and the accompanying drawings.

[0030] Please refer to Figure 1 The embodiment of the present application provides an optimal dispatch method of a distributed smart grid, comprising the following steps:

[0031] For each independent dispatch region of the distributed smart grid, a local distribution network optimal dispatch model corresponding to each independent dispatch region is established with the minimum operation cost of each independent dispatch region as the target, and the decision variables of the local distribution network optimal dispatch model include the output of local distributed power sources, the output of switchable capacitors and the position of on-load voltage regulating transformer taps;

[0032] The voltages on both sides of the tie line of adjacent independent dispatch regions are used as consistency variables, and the global distributed smart grid optimal dispatch model is constructed using the local distribution network optimal dispatch model;

[0033] The global distributed smart grid optimal dispatch model is converted into a standard alternating iteration problem;

[0034] The alternating direction multiplier method is used to solve the standard alternating iteration problem to obtain the dispatch strategy of the distributed smart grid.

[0035] From the above description, the beneficial effects of the present application are that: for each independent scheduling area of the distributed smart grid, a local power distribution network optimization scheduling model corresponding to each independent scheduling area is established with the minimum operation cost of each independent scheduling area as the target, then the voltage on both sides of the tie line of adjacent independent scheduling areas is taken as a consistency variable, the local power distribution network optimization scheduling model is used to build a global distributed smart grid optimization scheduling model, the standard alternating iteration problem is solved by using the alternating direction multiplier method to obtain the scheduling strategy of the distributed smart grid. Therefore, each independent area only interacts with the voltage information of the adjacent area, and the decoupling of each independent operation subject is realized through the tie line, which can well meet the privacy of the source and load information and the operation information of each independent operation subject. And the distributed power supply, the switchable capacitor, the on-load voltage regulating transformer tap and other smart grid regulation measures are considered in the present application, which can improve the potential of deep mining of the power grid and realize distributed optimization scheduling.

[0036] Further, the minimum operation cost of each independent scheduling area includes:

[0037] The objective function is established with the minimum operation cost as the target:

[0038]

[0039] In the formula, m represents the mth independent scheduling area, x m represents the decision variable of the local area, t represents the dynamic time, T represents the maximum time, l represents the lth line in the region, represents the purchase cost of the upper power supply, represents the operation cost of the distributed power supply, represents the operation cost of the switchable capacitor;

[0040]

[0041]

[0042]

[0043] In the formula, K gen represents the power supply transmission price of the upper power supply, K WTG,cut represents the unit cost of wind turbine abandoned electricity, K PVG,cut represents the unit cost of photovoltaic abandoned electricity, K CB represents the unit capacity of the switchable capacitor; represents the active power input by the upper power supply of the mth independent scheduling area at t time; represents the wind turbine abandoned active power of node j of the mth independent scheduling area at t time, represents the photovoltaic abandoned active power of node j of the mth independent scheduling area at t time, represents the switchable capacitor reactive power of the mth independent scheduling area node j at time t; Ω gen , Ω WTG , Ω PVG , Ω CB respectively represent the installation node set of the upper power supply point, the fan, the photovoltaic and the switchable capacitor in the mth independent scheduling area.

[0044] From the above description, according to the target function of the sum of the purchase cost of the upper power supply, the operation cost of the distributed power supply and the operation cost of the switchable capacitor being minimum, each independent operation subject can be fully considered, and the potential of the power grid can be deeply mined.

[0045] Further, the local power distribution network optimization scheduling model corresponding to each independent scheduling area is established, and further comprises:

[0046] establishing a branch power flow constraint condition, and performing second-order cone relaxation on the branch power flow constraint condition;

[0047] establishing a local power distribution network optimization scheduling model in combination with the target function and the relaxed branch power flow constraint condition.

[0048] From the above description, by performing second-order cone relaxation on the branch power flow constraint condition, the subsequent construction of the global distributed smart grid optimization scheduling model is facilitated.

[0049] Further, the constraint condition of the local power distribution network optimization scheduling model further comprises:

[0050] Distributed fan constraint:

[0051]

[0052] In the formula, represents the fan active power of the mth independent scheduling area node i at time t, represents the fan active power of the mth independent scheduling area node i at time t, represents the number of fans put into operation at time t, represents the maximum number of fans put into operation, s WTG represents the rated apparent power of the fan, represents the maximum reactive power of a single fan at time t, represents the fan reactive power of the mth independent scheduling area node i at time t;

[0053] Distributed photovoltaic power generation constraint:

[0054]

[0055] In the formula, This represents the photovoltaic active power of node i in the m-th independent scheduling region at time t. This represents the photovoltaic active power output of node i in the m-th independent scheduling region at time t. This represents the number of photovoltaic (PV) devices connected to node i in the m-th independent scheduling region at time t. Indicates the maximum amount of photovoltaic power invested, s PVG This indicates the rated apparent power of photovoltaic power. This represents the maximum reactive power output of a single photovoltaic unit at time t. Let i represent the photovoltaic reactive power of the m-th independent scheduling region node i at time t;

[0056] Transformer tap changer constraints:

[0057]

[0058]

[0059] In the formula, V represents the output voltage of the transformer tap at time t. seq This represents the adjustable voltage sequence of the transformer. A 0-1 auxiliary variable indicating whether the transformer tap changer is located at the upstream power source point at time t;

[0060] Switchable capacitor bank constraints:

[0061]

[0062]

[0063] In the formula, Let represent the reactive power of the switchable capacitor bank at time t for the m-th independent dispatch area node i. This represents the number of capacitor banks that can be switched on at time t for the m-th independent scheduling region node i. This indicates the rated capacity of a single capacitor in the switchable capacitor bank. This indicates the actual number of switchable capacitor banks installed at node i;

[0064] Represents the first binary auxiliary variable. Represents the second binary auxiliary variable. This represents the maximum number of adjustments that can be made to the switchable capacitor bank at node i. This represents the number of switchable capacitor banks connected at node i at time t-1.

[0065] From the above description, by setting the constraint condition, the smart grid adjustment measures such as the distributed power supply, the switchable capacitor, and the on-load voltage regulating transformer tap are fully considered, the potential of the power grid is deeply tapped, and the rationality of the distributed optimal scheduling is improved.

[0066] Further, the local power distribution network optimal scheduling model corresponding to each independent scheduling area is established, and further comprises:

[0067] The expression of setting the active power P injected by the node j is: j.t.m And the reactive power Q injected by the node j is: j.t.m

[0068]

[0069] In the formula, The active power and the reactive power input by the superior power supply are represented, The active power of the wind turbine and the photovoltaic is represented, The reactive power of the wind turbine, the photovoltaic and the switchable capacitor of the node j is represented respectively, The active power and the reactive power consumed by the load of the node j are represented.

[0070] Further, the voltage on both sides of the tie line of the adjacent independent scheduling area is taken as a consistency variable, and the global distributed smart grid optimal scheduling model is constructed using the local power distribution network optimal scheduling model, comprising:

[0071] The global distributed smart grid optimal scheduling model is established:

[0072]

[0073] s.t.x-z=0

[0074] x∈R x z∈R z

[0075] In the formula, M represents the total number of independent scheduling areas, x represents the decision variable of the global distributed smart grid, and z represents the consistency variable; R x represents the feasible region of the decision variable x, and R z represents the feasible region of the consistency variable z.

[0076] From the above description, in order to adapt to the distributed scheduling mode of the power grid-power grid, the global distributed scheduling model is established to ensure the effectiveness of subsequent calculation.

[0077] Further, the global distributed smart grid optimal scheduling model is converted into a standard alternating iteration problem, comprising:

[0078] The objective function of the global distributed smart grid optimal scheduling model is converted into:​

[0079]

[0080] From the above description, the global distributed smart grid optimization scheduling model is converted into a standard alternating iteration problem, so as to be solved by ADMM algorithm or improved algorithm subsequently.

[0081] Further, the standard alternating iteration problem is solved by using an alternating direction multiplier method, comprising:

[0082] initializing decision variables and scaled dual variables;

[0083] solving the mth independent scheduling region according to the initialized power flow distribution and the iteration voltage of the consistency variable on both sides of the tie line, and updating the decision variable of the mth independent scheduling region;

[0084] updating the semi-step scaled dual variable according to the solved decision variable, and updating the consistency variable of the mth independent scheduling region according to the voltage on both sides of the tie line and the semi-step scaled dual variable;

[0085] updating the scaled dual variable and the penalty function according to the semi-step scaled dual variable, the consistency variable and the decision variable;

[0086] judging whether the primal residual and the dual residual of the penalty function are converged or not, and if not, returning to perform the operation of solving the mth independent scheduling region according to the initialized power flow distribution and the voltage on both sides of the tie line until the primal residual and the dual residual of the penalty function are converged.

[0087] Further, the penalty function is:

[0088]

[0089] wherein k represents the iteration number, γ k represents the primal residual, γ k+1 =x k+1 -z k+1 ; s k represents the dual residual, s k+1 =ρ(z k+1 -z k ), x k represents the decision variable of the kth iteration, z k represents the consistency variable of the kth iteration.

[0090] From the above description, by improving the solving method of the alternating direction multiplier method, the scaled dual variable of the method is in a symmetric iteration mode, which can realize the fast convergence of the algorithm, greatly improve the calculation efficiency, and is suitable for engineering promotion.

[0091] Referring to Figure 2 Another embodiment of the present application provides an optimal scheduling terminal of a distributed smart grid, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements each step of the optimal scheduling method of the distributed smart grid when executing the computer program.

[0092] The optimal scheduling method and terminal of the distributed smart grid can improve the rationality of the distributed scheduling of the smart grid and the data privacy of the distributed subject, and the following specific embodiments are described.

[0093] Embodiment one

[0094] Referring to Figure 1 An optimal scheduling method of a distributed smart grid comprises the following steps:

[0095] S1, for each independent scheduling area of the distributed smart grid, a local power distribution network optimal scheduling model corresponding to each independent scheduling area is established with the minimum operation cost of each independent scheduling area as the target, and the decision variables of the local power distribution network optimal scheduling model include the output of the local distributed power source, the output of the switchable capacitor, and the position of the on-load voltage regulating transformer tap.

[0096] Specifically, the local optimal scheduling model is established, and for any independent operation area, a local optimal power flow (OPF) calculation model is constructed based on the distflow optimal power flow equation. The local OPF construction adopts the second order cone relaxing (SOCR) technology, and the minimum operation cost is taken as the objective function.

[0097] The target function is established with the minimum operation cost as the target, and the representation method is as follows:

[0098]

[0099] In the formula, m represents the mth independent scheduling area, x m represents the decision variable of the local area, t represents the dynamic time, T represents the maximum time, l represents the lth line in the area, represents the purchase cost of the upper-level power source, represents the operation cost of the distributed power source (including wind turbine WTG and photovoltaic PVG), represents the operation cost of the switchable capacitor;

[0100]

[0101]

[0102]

[0103] where K gen represents the transmission price of the upper-level power supply, K WTG,cut represents the unit cost of the wind turbine abandoned power, K PVG,cut represents the unit cost of the photovoltaic abandoned power, K CB represents the unit cost of the switchable capacitor, K represents the active power input by the upper-level power supply in the mth independent dispatching area at time t; represents the active power of the wind turbine abandoned power in the jth node of the mth independent dispatching area at time t, represents the active power of the photovoltaic abandoned power in the jth node of the mth independent dispatching area at time t, represents the reactive power of the switchable capacitor in the jth node of the mth independent dispatching area at time t; Ω gen , Ω WTG , Ω PVG , Ω CB respectively represent the installation node set of the upper-level power supply point, the wind turbine, the photovoltaic and the switchable capacitor in the mth area.

[0104] The branch power flow constraint of distflow is represented as follows:

[0105]

[0106] wherein subscript t represents the tth dynamic time, subscript m represents the mth local area, δ(j) represents the branch with node j as the first node, and π(j) represents the branch with node j as the last node;

[0107] P j.t.m , Q j.t.m respectively represent the injected active and reactive power of node j, P ij.t.m , Q ij.t.m respectively represent the active and reactive power flow of branch ij, R ij , X ij respectively represent the branch resistance and reactance. I ij.t.m represents the current of branch ij, V i.t.m represents the voltage of node i, I ij.min , I ij.max respectively represent the lower limit and upper limit of the current of branch ij, V i.min , V i.max respectively represent the lower limit and upper limit of the voltage of node i.

[0108] The results after the second-order cone relaxation are as follows:

[0109]

[0110] wherein, denotes the square of branch current ij, denotes the square of node voltage vi, denotes the lower limit and upper limit of branch current square, respectively, denotes the lower limit and upper limit of node voltage square, respectively.

[0111] Further, the established local power distribution network optimal power flow model also includes the following constraints:

[0112] (1) Establishing distributed wind turbine constraints:

[0113]

[0114] wherein, denotes the wind turbine active power of node i in the mth independent dispatching area at time t, denotes the wind turbine active output of node i in the mth independent dispatching area at time t, denotes the number of wind turbines put into operation of node i in the mth independent dispatching area at time t, denotes the maximum number of wind turbines put into operation, s WTG denotes the rated apparent power of the wind turbine, denotes the maximum reactive output of a single wind turbine at time t, denotes the wind turbine reactive power of node i in the mth independent dispatching area at time t.

[0115] (2) Establishing distributed photovoltaic power generation constraints:

[0116]

[0117] wherein, denotes the photovoltaic active power of node i in the mth independent dispatching area at time t, denotes the photovoltaic active output of node i in the mth independent dispatching area at time t, denotes the number of photovoltaics put into operation of node i in the mth independent dispatching area at time t, denotes the maximum number of photovoltaics put into operation, s PVG denotes the rated apparent power of the photovoltaic, denotes the maximum reactive output of a single photovoltaic at time t, denotes the photovoltaic reactive power of node i in the mth independent dispatching area at time t.

[0118] (3) Establishing transformer tap constraints:

[0119]

[0120]

[0121] In the formula, Vt represents the output voltage of the transformer tap at time t, seq Vt represents the adjustable voltage sequence of the transformer, 0-1 auxiliary variable, the transformer tap is located at the upper power supply point, considering the characteristics of open-loop operation of the power distribution network, only the local area of the upper power supply point exists transformer tap constraint.

[0122] (4) Establish the switchable capacitor bank constraint:

[0123]

[0124]

[0125] In the formula, Qm represents the reactive power of the switchable capacitor bank of the mth independent dispatching area node i at time t, Nt represents the number of switchable capacitor banks of the mth independent dispatching area node i at time t, C represents the rated capacity of a single capacitor of the switchable capacitor bank, N represents the actual installation number of the switchable capacitor bank of node i;

[0126] Y1 represents the first binary auxiliary variable, Y2 represents the second binary auxiliary variable, Nmax represents the maximum number of adjustments of the switchable capacitor bank of node i, Nt-1 represents the number of switchable capacitor banks of node i at time t-1.

[0127] After fully considering various active management elements, the power injection expression of node j is as follows:

[0128]

[0129] In the formula, Pj represents the active power and reactive power input by the upper power supply, Pf and Pv represent the active power of the fan and the photovoltaic, Qf, Qv and Qc represent the reactive power of the fan, the photovoltaic and the switchable capacitor of node j respectively, Pj and Qj represent the active power and reactive power consumed by the load of node j.

[0130] S2, taking the voltages on both sides of the tie line of adjacent independent dispatching areas as consistency variables, using the local power distribution network optimization scheduling model to construct a global distributed smart grid optimization scheduling model.

[0131] Wherein, since the micro-grid in the embodiment is not a point load, but a meshed area, there are multiple dispatching mechanisms within the area range, each of which is responsible for its own dispatching behavior, that is, each dispatching mechanism is only responsible for the power flow within its own jurisdiction. The multiple dispatching centers distribute the dispatching through the voltage on both sides of the tie line, and need to further build a global distributed power flow calculation model.

[0132] Specifically, the global distributed smart grid optimization dispatching model is established:

[0133]

[0134] s.t.x-z=0

[0135] x∈R x ,z∈R z

[0136] In the formula, M represents the total number of independent dispatching areas, x represents the decision variable of the global distributed smart grid, which is composed of x1, x2, …, xM, and z represents the consistency variable; R m Each local area decision variable constitutes, z represents the consistency variable; R x represents the feasible region of the decision variable x, and R z represents the feasible region of the consistency variable z.

[0137] S3, converting the global distributed smart grid optimization dispatching model into a standard alternating iteration problem.

[0138] Considering the constraint x-z=0, in the iteration process, if the problem converges, when x→x*, z→z*, x*=z*, wherein x*, z* respectively represent the optimal solution, so the global distributed smart grid optimization dispatching model in step S2 can be equivalent to a standard dual ascent problem, as follows:

[0139]

[0140] s.t. x-z=0

[0141] x∈R x ,z∈R z

[0142] It can be further equivalent to:

[0143]

[0144] S4, using the alternating direction multiplier method to solve the standard alternating iteration problem to obtain the dispatching strategy of the distributed smart grid.

[0145] S41, initializing the decision variable x 0 , z 0and the scale dual variable μ 0 The superscript here in this embodiment indicates the iteration number.

[0146] S42, according to the initialization flow distribution, the voltage of the kth iteration result of the consistency variable z on both sides of the tie line is updated according to the decision variable x of the mth independent scheduling area and μ k Substitute the following formula, solve the mth independent scheduling area in turn, and update the decision variable x of the mth independent scheduling area in turn m :

[0147]

[0148] In the formula, k represents the kth iteration, when k=1, the initial value is brought in; The voltage on both sides of the tie line is represented by ρ, and the penalty function is represented by ρ.

[0149] S43, update the semi-step scale dual variable μ according to the decision variable obtained by solving k+0.5 :

[0150]

[0151] S44, substitute and μ k+0.5 into the following formula, and update the consistency variable z of the mth independent area in turn m :

[0152]

[0153] S45, update the scale dual variable μ k+1 .

[0154]

[0155] S46, update the penalty function ρ:

[0156]

[0157] In the formula, k represents the iteration number, γ k represents the original residual error, γ k+1 =x k+1 -z k+1 ; s k represents the dual residual error, s k+1 =ρ(z k+1 -z k ), x k represents the decision variable of the kth iteration, z k represents the consistency variable of the kth iteration.

[0158] S47, repeat steps S42-S46 until the algorithm converges, wherein the convergence condition is as follows:

[0159]

[0160] In the formula, ε pri and ε dual This is a preset, relatively small value, typically on the order of 10. -3 Or 10 -4 .

[0161] Example 2

[0162] To demonstrate the superiority of the planning model described in this embodiment, this embodiment employs the following... Figure 3 The 3-zone IEEE 33-bus distribution system shown was simulated and verified. Figure 3 Node 0 is the upstream power supply node, and it divides the power distribution system into three independent operating areas. See the table below for details:

[0163] Table 1 Distribution Network Zoning and Parameters

[0164]

[0165] Region 1 is connected to Region 2 and Region 4 via Line 6-7 and Line 5-25 respectively. Therefore, this embodiment has two connecting lines, namely Line 6-7 and Line 5-25.

[0166] In this embodiment, residential load nodes are nodes 1-17; commercial load nodes are nodes 25-32; and industrial load nodes are nodes 18-24. The system reference load is twice the reference value of the standard IEEE 33-node distribution system, that is, in the model proposed in this embodiment, the reference load value is 10168.52kW + 5094.64kVar. The load characteristic curves for residential, commercial, and industrial loads are shown below. Figures 4 to 6 As shown, the power output curves of wind turbines and photovoltaic power are as follows: Figure 7 and Figure 8 As shown, where Figures 4 to 8 The spring curve is consistent with the autumn curve.

[0167] The parameters for this embodiment are explained as follows: the electricity purchase price from the upstream power source is 0.42 yuan / kWh; the unit cost of DG curtailment is 2 yuan / kWh; the unit capacity CB commissioning cost is 0.1 yuan / kVar; the rated power factor of WTG is 0.9; the rated power factor of PVG is 1; the OLTC has 9 adjustment levels, with a voltage adjustment step size of 0.0125pu; the adjustable voltage range is 0.95pu~1.05pu; the maximum allowable voltage deviation of the distribution system is ±7%; the unit capacity of WTG is 0.3MVA; the unit capacity of PVG is 0.3MVA; the capacity of each CB group is 0.1Mvar, and the total number of daily adjustments for capacitors does not exceed 30 groups.

[0168] The configuration of various devices of the distributed smart grid is shown in the following table. The WTG is installed at nodes 22 and 32, 3 groups and 5 groups respectively, the PVG is installed at nodes 11 and 26, 5 groups respectively, and the CB is installed at node 25, 10 groups.

[0169] Table 2 Configuration of various devices of the distributed smart grid

[0170]

[0171] The active and reactive power output of various power sources in the embodiment is shown in Figure 9 and Figure 10 The action of the switchable capacitor is shown in Figure 11 The total annual operation cost of the distributed smart grid is 1094.58 million yuan. Through simulation, it is found that the distributed algorithm proposed in the embodiment is effective.

[0172] Embodiment Three

[0173] Please refer to Figure 2 An optimal scheduling terminal of a distributed smart grid 1, comprising a memory 2, a processor 3, and a computer program stored in the memory 2 and executable on the processor 3, wherein the processor 3 implements each step of the optimal scheduling method of the distributed smart grid of embodiment one or two when executing the computer program.

[0174] In summary, the optimal scheduling method and terminal of the distributed smart grid provided by the present application can realize the decoupling of each independent operation subject through the interconnection line, and can well meet the privacy of the source and load information and the operation information of each independent operation subject. In addition, the present application fully considers the smart grid adjustment measures such as distributed power, switchable capacitor, and on-load voltage regulating transformer tap, deeply excavates the potential of the power grid, and realizes the distributed optimal scheduling. At the same time, the present application proposes an improved solution method of the alternating direction multiplier method, the scale pair variable of the proposed method adopts a symmetric iteration method, which can realize the fast convergence of the algorithm, greatly improves the calculation efficiency, and is suitable for engineering promotion.

[0175] The above description is only an embodiment of the present application, and does not limit the patent range of the present application, and any equivalent transformation or direct or indirect application in related technical fields using the content of the specification and drawings of the present application is also included in the patent protection range of the present application.

Claims

1. An optimized scheduling method for a distributed smart grid, characterized in that, Including the following steps: For each independent dispatch area of ​​the distributed smart grid, a local distribution network optimization dispatch model is established for each independent dispatch area with the goal of minimizing the operating cost of each independent dispatch area. The decision variables of the local distribution network optimization dispatch model include the output of local distributed power sources, the output of switchable capacitors, and the location of the tap changer of the on-load tap-changing transformer. Using the voltages on both sides of the tie line between adjacent independent dispatch areas as consistency variables, a global distributed smart grid optimization dispatch model is constructed using the local distribution network optimization dispatch model. The global distributed smart grid optimization scheduling model is transformed into a standard alternating iterative problem. The standard alternating iterative problem is solved using the alternating direction multiplier method to obtain the scheduling strategy of the distributed smart grid. The objective of minimizing the operating cost of each independent scheduling region includes: The objective function is established with the goal of minimizing operating costs: In the formula, m Indicates the first m Each independent scheduling area x m Represents decision variables for this region. t Indicates a dynamic moment. T Indicates the maximum moment. l Indicates the first in this region l One line, This indicates the cost of electricity purchased from the upstream power source. This represents the operating cost of distributed power sources. This indicates the operating cost of switchable capacitors; In the formula, K gen This indicates the transmission price of electricity from the upstream power source. K WTG,cut This indicates the unit cost of wind turbine power curtailment. K PVG,cut This indicates the unit cost of photovoltaic curtailment. K CB This indicates the commissioning cost per unit capacity of switchable capacitors; express t Time of the first m The active power input from the upstream power source of each independent dispatch area; express t Time of the first m Each independent scheduling area node j The active power of wind turbine curtailment express t Time of the first m Each independent scheduling area node j The active power of curtailed photovoltaic power, express t Time of the first m Each independent scheduling area node j Reactive power of switchable capacitors; Ω gen Ω WTG Ω PVG Ω CB They represent the first time. m A collection of installation nodes for upstream power sources, wind turbines, photovoltaic systems, and switchable capacitors within an independent dispatch area; Using the voltages on both sides of the tie line between adjacent independent dispatch areas as consistency variables, a global distributed smart grid optimization dispatch model is constructed using the local distribution network optimization dispatch model, including: Establish the global distributed smart grid optimal scheduling model: In the formula, M This indicates the total number of independently scheduled regions. x Represents the decision variables of a globally distributed smart grid. z Representing consistency variables; R x Representing decision variables x feasible domain, R z Representing consistency variables z The feasible domain.

2. The optimized scheduling method for a distributed smart grid according to claim 1, characterized in that, Establishing local distribution network optimization scheduling models for each independent scheduling area also includes: Establish branch flow constraints and perform second-order cone relaxation on the branch flow constraints; A local distribution network optimization scheduling model is established by combining the objective function and the relaxed branch power flow constraints.

3. The optimized scheduling method for a distributed smart grid according to claim 1, characterized in that, The constraints of the local distribution network optimization scheduling model also include: Distributed wind turbine constraints: In the formula, express t Time of the first m Each independent scheduling area node i The active power of the wind turbine, express t Time of the first m Each independent scheduling area node i The wind turbine has power output. express t Time of the first m Each independent scheduling area node i The number of wind turbines put into operation Indicates the maximum number of wind turbines that can be put into operation. s WTG This indicates the rated apparent power of the fan. express t The maximum reactive power output of a single wind turbine at any given time. express t Time of the first m Each independent scheduling area node i The reactive power of the wind turbine; Constraints of distributed photovoltaic power generation: In the formula, express t Time of the first m Each independent scheduling area node i Photovoltaic active power, express t Time of the first m Each independent scheduling area node i The photovoltaic system is contributing its power. express t Time of the first m Each independent scheduling area node i The number of photovoltaic investments, Indicates the maximum amount of photovoltaic power invested. s PVG This indicates the rated apparent power of photovoltaic power. express t The maximum reactive power output of a single photovoltaic unit at any given time. express t Time of the first m Each independent scheduling area node i Photovoltaic reactive power; Transformer tap changer constraints: In the formula, express t The output voltage of the transformer tap changer at all times. V seq This represents the adjustable voltage sequence of the transformer. express t A 0-1 auxiliary variable indicating whether the transformer tap changer is located at the upstream power supply point; Switchable capacitor bank constraints: In the formula, express t Time of the first m Each independent scheduling area node i The reactive power of switchable capacitor banks, express t Time of the first m Each independent scheduling area node i The number of switchable capacitor banks that can be connected. This indicates the rated capacity of a single capacitor in the switchable capacitor bank. Represents a node i The actual number of switchable capacitor banks installed; Represents the first binary auxiliary variable. Represents the second binary auxiliary variable. Represents a node i The maximum number of adjustments that can be made to the switchable capacitor bank. express t The node at time -1 i The number of switchable capacitor banks to be connected.

4. The optimized scheduling method for a distributed smart grid according to claim 3, characterized in that, Establishing local distribution network optimization scheduling models for each independent scheduling area also includes: Set up nodes j Injected active power P j.t.m and nodes j Injected reactive power Q j.t.m The expression: In the formula, , This indicates the active and reactive power input from the upstream power source. , This indicates the active power of wind turbines and photovoltaic systems. , , Representing nodes respectively j Reactive power of wind turbines, photovoltaic systems, and switchable capacitors. , For nodes j The active and reactive power consumed by the load.

5. The optimized scheduling method for a distributed smart grid according to claim 1, characterized in that, The global distributed smart grid optimization scheduling model is transformed into a standard alternating iterative problem, including: The objective function of the global distributed smart grid optimization scheduling model is transformed into: 。 6. The optimized scheduling method for a distributed smart grid according to claim 1, characterized in that, The standard alternating iterative problem is solved using the alternating direction multiplier method, including: Initialize the decision variables and scaling even variables; Based on the initial power flow distribution and the iterative voltages of the consistency variables on both sides of the tie line, the first... m Solve the problem in each independent scheduling region, and update the [number]th [item] sequentially. m Decision variables for each independent scheduling region; Update the half-step scaling even variables based on the decision variables obtained from the solution, and update the first half-step scaling even variables sequentially based on the voltages on both sides of the tie line and the half-step scaling even variables. m Consistent variables for each independent scheduling region; Update the scaling even variable and penalty function based on the half-step scaling even variable, consistency variable, and decision variable; Determine whether both the original residual and the dual residual of the penalty function have converged. If not, return to the previous step and proceed according to the initial power flow distribution and the voltages on both sides of the tie line, sequentially applying the penalty function to the first... m The solution operation is performed in an independent scheduling region until both the original residual and the dual residual of the penalty function converge.

7. The optimized scheduling method for a distributed smart grid according to claim 6, characterized in that, The penalty function is: In the formula, k Indicates the number of iterations. Represents the original residual. ; s k Represents the dual residual. , x k Indicates the first k Decision variables for the next iteration z k Indicates the first k Consistency variables for each iteration.

8. An optimized dispatch terminal for a distributed smart grid, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements each step of the optimized scheduling method for a distributed smart grid as described in any one of claims 1 to 7.

Images