Coordinated voltage-reactive power dual-layer optimization method and device for CVR regulation equipment and inverter

Abstract

The present invention relates to a voltage-reactive power dual-layer optimization method and device for coordinating CVR regulation equipment and inverters. The method comprises the following steps: first, establishing a three-phase power flow model based on branch power flow; second, establishing an approximate three-phase power flow model and a linearized three-phase power flow model based on the branch power flow based on the three-phase power flow model of the branch power flow; third, respectively modeling the voltage regulator, capacitor bank, inverter, and voltage-sensitive load; fourth, using the linearized three-phase power flow model to determine the setting values of the traditional CVR regulation equipment and inverter to establish an upper-layer MILP model, and correcting the setting value of the inverter based on the approximate nonlinear optimal power flow model to obtain an approximate optimal feasible solution and a lower-layer NLP model; solving the upper-layer MILP model and the lower-layer NLP model, and performing voltage-reactive power dual-layer optimization based on the results. The present invention can reduce user electricity demand, reduce complexity, and achieve scalability, and has good scalability and high efficiency.

Description

Technical Field

[0001] The present invention relates to the field of reactive power optimization and energy saving of power distribution systems, and more particularly to a voltage-reactive power double-layer optimization method and device for coordinating CVR regulation equipment and inverters. Background Art

[0002] Conservation voltage reduction (CVR) is a technology that reduces user power demand through feeder voltage regulation, thereby achieving energy conservation. Feeder voltage regulation is based on the load's sensitivity to feeder voltage. By regulating voltage, it reduces overall losses in the distribution system while also reducing the amount of power required by the load.

[0003] Currently, distribution system CVR equipment primarily utilizes two types of reactive power regulation devices: on-load tap regulators (OLTs), which primarily regulate feeder voltage by adjusting their taps; and capacitor banks, whose primary purpose is to provide sufficient reactive power for the distribution system and reduce losses. However, with the widespread grid connection of inverter-based distributed power sources, energy storage, and other devices, feeder voltage regulation in distribution systems has become increasingly diverse. Typical CVR methods typically utilize devices such as OLTs and capacitor banks. These reactive power regulation devices all have a certain time delay, representing a slow-time-scale regulation. However, for a large number of grid-connected distributed power sources and energy storage systems, their inverter switching is more frequent and rapid, representing a fast-time-scale regulation. However, existing CVR technologies lack coordination between inverters and traditional CVR regulation devices, and also lack consideration of timescales. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a voltage-reactive dual-layer optimization method and device for coordinating CVR regulation equipment and inverters; the method can achieve coordination on a time scale, and achieve energy-saving benefits of a three-phase power distribution system by simultaneously coordinating traditional CVR regulation equipment and inverters, thereby reducing user electricity demand and improving energy-saving effects.

[0005] The technical solution adopted by the present invention to solve the technical problem is to construct a voltage-reactive power dual-layer optimization method for coordinating CVR regulation equipment and inverter, including the following steps:

[0006] The following steps are involved:

[0007] The first step is to establish a three-phase power flow model based on branch power flow;

[0008] The second step is to establish an approximate three-phase power flow model and a linearized three-phase power flow model based on the branch power flow according to the three-phase power flow model of the branch power flow;

[0009] The third step is to model the voltage regulator, capacitor bank, inverter and voltage-sensitive load respectively;

[0010] In the fourth step, a linearized three-phase power flow model is used to determine the setting values of the traditional CVR regulation equipment and the inverter to establish an upper-level MILP model, and the setting values of the inverter are corrected according to the approximate nonlinear optimal power flow model to obtain an approximate optimal feasible solution and a lower-level NLP model; the upper-level MILP model and the lower-level NLP model are solved, and voltage-reactive power double-level optimization is performed based on the results.

[0011] According to the above scheme, the three-phase power flow model based on branch power flow is:

[0012]

[0013] Rank-1 positive semidefinite (PSD) matrix

[0014] Where, is the complex power of phase p, is the p-phase complex voltage, is the p-phase current, Φ i represents the phase set of busbar i, z ij is the phase sequence impedance matrix, G = (N, ε), where N represents the busbar set, ε represents the branch set, each branch (i, j) represents the connection form between two adjacent nodes i and j, {a, b, c} represents the three phases of the system, for each busbar;

[0015] For p∈Φ i Phase, complex power demanded by the load at bus i, and are active and reactive power demands respectively; (pq)∈Φ ij The complex power of branch (i, j) is and Corresponding to active power and reactive power respectively; is the complex impedance matrix z of branch (i, j) ij An element of , where (pq): p∈Φ i ,q∈Φ j ;

[0016] In the above formula, the first formula represents the voltage drop equation, the second formula represents the power balance equation, the third formula represents the power equation calculation formula, and the fourth formula is the rank 1 constraint.

[0017] According to the above scheme, the approximate three-phase power flow model based on branch power flow is:

[0018]

[0019] Among them, for all (i, j)∈ε, the first and second formulas represent pp∈Φ ij The branch (i, j) active power and reactive power flow equations, and Assumed to be a constant, the first and second equations are and is linear; the third formula is the voltage drop equation between the two nodes of the (i, j) branch corresponding to p, In the equation, the third equation is linear; the fourth equation relates the complex power flow of each phase in branch (i, j) to the phase voltage and phase current, and the fourth equation is a nonlinear quadratic equality constraint;

[0020] where p∈Φ i ; where (pq)∈Φ ij ; is the branch current and The phase difference between

[0021] The phase voltages differ by approximately 120°, which satisfies:

[0022]

[0023] is a constant, S ij The elements of are approximated by diagonal moments;

[0024] The phase currents of phases p and q are and Given busbar i, The phase difference of the upper phase current is approximately a constant variable. By solving the equivalent distribution system flow when the load is a constant impedance model, the branch (i, j) is obtained. approximation;

[0025] The linearized three-phase power flow model is:

[0026]

[0027] The model of the voltage regulator is:

[0028] and

[0029] a p is the turns ratio of the voltage regulator connected to the (i, j) branch p, that is, a pThe value of is between 0.9 and 1.1, and each step produces a change of 0.00625pu. An additional node i′ is introduced. For each step position of the voltage regulator, that is, i∈(1, 2, ..., 32), For a 0-1 variable, define a vector b i ∈{0.9, 0.90625, ..., 1.1}, then V i p , V j p , and

[0030] Where p∈Φ i ∩Φ j as follows:

[0031] and

[0032] in,

[0033] and

[0034] To express the above as and function, take the square of the above formula, and define and and

[0035] According to the above scheme, the capacitor model is:

[0036]

[0037] in, Indicates the on / off status of the capacitor bank, is the rated single-phase reactive power of the capacitor bank, is the square of the voltage of busbar i of phase p, is the reactive power generated by the capacitor bank.

[0038] According to the above scheme, the inverter model is:

[0039]

[0040] is the available reactive power of the inverter. The known active power generation of DG is equal to the predicted value. The reactive power output of DG depends on the rated value of the inverter. To connect i th The rated single-phase apparent power capacity of the inverter at DG, To predict active power generation.

[0041] According to the above scheme, the specific method of voltage-sensitive load modeling is as follows:

[0042] First, the mathematical model of the ZIP model of the load connected to the p-phase of bus i is established as follows:

[0043]

[0044] Among them, k p,1 +k p,2 +k p,3 =1,k q,1 +k q,2 +k q,3 =1, and is the active power demand and reactive power demand of single-phase load under standard voltage;

[0045] Then, the ZIP load model is transformed into an equivalent load model using the CVR coefficient. The CVR coefficient is defined as the ratio of the percentage of active or reactive power reduction to the percentage of bus voltage reduction. The CVR coefficient of active and reactive power reduction is CVR. p 、CVR q They are:

[0046]

[0047] in, and therefore Let V i p ≈V0 and It turns out that:

[0048]

[0049] The above load model is based on the CVR coefficient A linear load model based on the CVR coefficient Embedded into the three-phase power flow model established in the second step; CVR coefficient CVR p and CVR q It can be estimated by the ZIP coefficient of the load, differentiating the ZIP model and assuming V0 = 1 p.u., to obtain:

[0050]

[0051] Assume V i p ≈V0, the CVR coefficient of the load is estimated by the ZIP coefficient:

[0052] CVR p =2k p,1+k p,2 、CVR q =2k q,1 +k q,2 .

[0053] According to the above scheme, the specific method of using the linearized three-phase power flow model to determine the setting values of the traditional CVR regulation equipment and inverter to establish the upper MILP model is as follows:

[0054] By controlling the voltage regulator, capacitor bank and inverter to minimize the system loss, the optimization model is as follows to minimize the sum of the three-phase active power injected into the substation bus:

[0055] Objective function:

[0056] The constraints are as follows:

[0057] Linear AC power flow constraints:

[0058]

[0059] CVR-based load model:

[0060]

[0061]

[0062] Voltage regulator model:

[0063]

[0064] Capacitor Bank Model:

[0065] Inverter model:

[0066]

[0067] Voltage Constraints:

[0068] The variables in the above model are It is a mixed integer programming model.

[0069] According to the above scheme, the inverter setting value is corrected according to the approximate nonlinear optimal power flow model to obtain an approximate optimal feasible solution. The specific method for obtaining the lower-level NLP model is as follows:

[0070] Adjust the setting value of the inverter control variable to obtain the optimal feasible three-phase nonlinear power flow solution, assuming that the discrete control variable For fixed, a real-time correction model of the inverter setting value based on the approximate nonlinear optimal power flow model is established as follows:

[0071] Objective function:

[0072] The constraints are as follows:

[0073] The constraints of the approximate nonlinear AC power flow equation defined at time t are:

[0074]

[0075]

[0076] CVR-based load model:

[0077]

[0078] Voltage regulator model: Capacitor Bank Model: Inverter model:

[0079]

[0080] Voltage Constraints:

[0081] The variables in the above model are The optimal control setpoints for inverter reactive power dispatch are obtained for a nonlinear optimization model with linear objectives and quadratic constraints.

[0082] The present invention also provides a voltage-reactive power dual-layer optimization device for coordinating a CVR regulating device and an inverter, comprising:

[0083] Three-phase power flow model establishment module, used to establish a three-phase power flow model based on branch power flow;

[0084] An approximate three-phase power flow model and a linearized three-phase power flow model establishing module, configured to establish an approximate three-phase power flow model and a linearized three-phase power flow model based on branch power flow according to the three-phase power flow model based on branch power flow;

[0085] Model building modules for modeling voltage regulators, capacitor banks, inverters, and voltage-sensitive loads separately;

[0086] The voltage-reactive power dual-layer optimization module is used to use a linearized three-phase power flow model to determine the setting values of traditional CVR regulation equipment and inverters to establish an upper-layer MILP model, and to correct the inverter setting values according to the approximate nonlinear optimal power flow model to obtain an approximate optimal feasible solution and a lower-layer NLP model; the upper-layer MILP model and the lower-layer NLP model are solved, and the voltage-reactive power dual-layer optimization is performed based on the results.

[0087] The implementation of the voltage-reactive dual-layer optimization method of the coordinated CVR regulation device and inverter of the present invention has the following beneficial effects:

[0088] 1. The present invention develops an approximate model and a linear model of the three-phase power distribution system power flow model, which is applicable to the three-phase unbalanced power distribution system. Compared with the strict three-phase power distribution system power flow model, the approximate model reduces the dimension of the power flow variables and has high approximation accuracy.

[0089] 2. The present invention converts the ZIP load model into a load model based on the CVR coefficient, which is convenient for embedding into the three-phase distribution system power flow model without changing its original structure;

[0090] 3. Based on the decomposition and coordination idea, the present invention decomposes the MINLP problem of CVR regulation equipment and inverter coordination into a two-level optimization problem of MILP and NLP, which can reduce complexity and achieve scalability, and has good scalability and high efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0091] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0092] Figure 1 The present invention is a flow chart of a voltage-reactive power dual-layer optimization method for coordinating a CVR regulating device and an inverter. DETAILED DESCRIPTION

[0093] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, specific embodiments of the present invention are now described in detail with reference to the accompanying drawings.

[0094] like Figure 1 As shown, the voltage-reactive dual-layer optimization method of the coordinated CVR regulation device and the inverter of the present invention includes the following steps:

[0095] The first step is to establish a three-phase power flow model based on branch power flow;

[0096] Assume a directed graph G = (N, ε), where N represents the busbar set and ε represents the branch set. Each branch (i, j) represents the connection between two adjacent nodes i and j. {a, b, c} represents the three phases of the system, Φ i represents the phase set of bus i. For each bus i∈N, the p-phase complex voltage is assumed to be V i p , the complex power of phase p is assumed to be make For each branch, let the p-phase current be and define Assume z ij is the phase sequence impedance matrix. The radial three-phase distribution system power flow model based on branch power flow is as follows:

[0097]

[0098] Rank-1 positive semidefinite (PSD) matrix

[0099] Where, For p∈Φ i Phase, complex power demanded by the load at bus i, and are active and reactive power demands respectively; (pq)∈Φ ij The complex power of branch (i, j) is and Corresponding to active power and reactive power respectively; is the complex impedance matrix z of branch (i, j ij An element of , where (pq): p∈Φ i ,q∈Φ j .

[0100] In the above formula, the first formula represents the voltage drop equation; the second formula represents the power balance equation; the third formula represents the power equation calculation formula; and the fourth formula is the rank 1 constraint.

[0101] The second step is to establish an approximate three-phase power flow model and a linearized three-phase power flow model based on the branch power flow according to the three-phase power flow model of the branch power flow;

[0102] definition where p∈Φ i ; where (pq)∈Φ ij ; Notice, is the branch current and The phase difference between them.

[0103] First, assume that the phase voltages differ by approximately 120°, that is,

[0104]

[0105] further

[0106]

[0107] Using the above two formulas, we can Expressed as The function of is a constant. ij The elements of can be approximated by diagonal moments, which helps to reduce the number of power flow variables.

[0108] Then, assume that for a given line (i, j), the phase currents of phases p and q are and Then given busbar i, The phase difference of the upper phase current is approximately a constant variable. By solving the equivalent distribution system power flow when the load is a constant impedance model, the branch (i, j) is obtained. approximation.

[0109] Finally, using the above assumptions, the three-phase power flow model based on branch power flows established in the first step is approximated as a set of linear and nonlinear equations as shown below (approximate three-phase power flow model based on branch power flows):

[0110]

[0111] Among them, for all (i, j)∈ε, the first and second formulas represent pp∈Φ ij The active power and reactive power flow equations of the branch (i, j). and is assumed to be a constant, so the equation and The third equation is the voltage drop equation between the two nodes of the (i, j) branch corresponding to p. The fourth equation relates the complex power flow of each phase in branch (i, j) to the phase voltage and phase current, which is a nonlinear quadratic equality constraint.

[0112] If the branch power loss is assumed to be negligible, the approximate three-phase power flow model based on branch power flow can be linearized as (Linearized three-phase power flow model based on branch power flow)

[0113]

[0114] The third step is to model the voltage regulator, capacitor bank, inverter and voltage-sensitive load respectively;

[0115] The voltage regulator modeling method is as follows:

[0116] Assume a 32-step voltage regulator with a voltage regulation range of ±10%. p is the turns ratio of the voltage regulator connected to the (i, j) branch p, that is, a p The value of is between 0.9 and 1.1, and each step produces a change of 0.00625 pu. An additional node i′ is introduced. For each step position of the voltage regulator, that is, i∈(1, 2, ..., 32), Is a 0-1 variable, define a vector b i ∈{0.9, 0.90625, ..., 1.1}. Then Vi p , V j p , and where p∈Φ i ∩Φ j as follows:

[0117] and

[0118] in, and

[0119] To express the above as and function, take the square of the above formula, and define and and The model describing the voltage regulator can be described as

[0120] and

[0121] The capacitor bank modeling method is as follows:

[0122] Reactive power generated by the capacitor bank is defined as a function of a binary control variable, namely

[0123]

[0124] in, Indicates the state of the capacitor bank (on / off), and its rated single-phase reactive power is and the square of the voltage of busbar i of phase p

[0125] The inverter modeling method is as follows:

[0126] The known active power generation of DG is equal to the predicted value. The reactive power output of DG depends on the rated value of the inverter. th The rated single-phase apparent power capacity of the inverter at DG is The predicted active power generation is Available reactive power of the inverter It is given by:

[0127]

[0128] The voltage-sensitive load modeling method is as follows:

[0129] First, the mathematical model of the ZIP model of the load connected to the p-phase of bus i is established as follows:

[0130]

[0131] Among them, k p，1 +k p，2 +k p，3 =1,k q，1 +k q，2 +k q，3 =1, and It is the active power demand and reactive power demand of single-phase load under standard voltage.

[0132] Then, the present invention uses the CVR coefficient to transform the ZIP load model into an equivalent load model. The CVR coefficient is defined as the ratio of the percentage of active or reactive power reduction to the percentage of bus voltage reduction. The CVR coefficient of active and reactive power reduction is CVR. p 、CVR q They are:

[0133]

[0134] in, and therefore Let V i p ≈V0 and It can be concluded that:

[0135]

[0136] Note that the above formula is based on the load model of the CVR coefficient It is linear and can therefore be easily embedded into the three-phase power flow model established in the second step. p and CVR q It can be estimated by the ZIP coefficient of the load. Differentiating the ZIP model and assuming V0 = 1 p.u., we can get:

[0137]

[0138] Assume V i p ≈V0, the CVR coefficient of the load can be estimated by the ZIP coefficient

[0139] CVR p =2k p，1 +k p，2 、CVR q =2k q，1 +k q，2 .

[0140] S4. Use a linearized three-phase power flow model to determine the setting values of the traditional CVR regulation equipment and the inverter to establish an upper-level MILP model, and correct the setting values of the inverter according to the approximate nonlinear optimal power flow model to obtain an approximate optimal feasible solution and a lower-level NLP model; solve the upper-level MILP model and the lower-level NLP model, and perform voltage-reactive power double-layer optimization based on the results.

[0141] The upper layer uses a linearized three-phase optimal power flow model at a scale of 14.5-15.5 minutes, preferably 15 minutes, to determine the setting values of the traditional CVR regulation equipment and the inverter. The lower layer uses an approximate nonlinear optimal power flow model at a scale of 0.9-1.1 minutes, preferably 1 minute, to correct the setting values of the inverter to obtain a near-optimal feasible solution. This embodiment uses the Gurobi toolbox and the fmincon function on the Mtalab platform to solve the upper-layer MILP model and the lower-layer NLP model, respectively, and outputs the results.

[0142] The method of establishing a coordinated optimization model based on the linearized three-phase optimal power flow model at the upper layer is:

[0143] The goal at this stage is to minimize system losses by controlling the voltage regulator, capacitor bank, and inverter, so as to minimize the sum of the three-phase active power injected into the substation bus. The optimization model is as follows:

[0144] Objective function:

[0145] The constraints are as follows:

[0146] Linear AC power flow constraints:

[0147]

[0148]

[0149] CVR-based load model:

[0150]

[0151] Voltage regulator model:

[0152]

[0153] Capacitor Bank Model:

[0154] Inverter model:

[0155]

[0156] Voltage Constraints:

[0157] The variables in the above model are It is a typical mixed integer programming model.

[0158] The method for establishing a real-time correction model for inverter setting values based on an approximate nonlinear optimal power flow model at the lower level at the min scale is:

[0159] The goal of this stage is to adjust the setting values of the inverter control variables to obtain the optimal feasible three-phase nonlinear power flow solution. The real-time correction model of the inverter setting value based on the approximate nonlinear optimal power flow model is established as follows:

[0160] Objective function:

[0161] The constraints are as follows:

[0162] The constraints of the approximate nonlinear AC power flow equation defined at time t are:

[0163]

[0164] CVR-based load model:

[0165]

[0166]

[0167] Voltage regulator model:

[0168] Capacitor Bank Model:

[0169] Inverter model:

[0170]

[0171] Voltage Constraints:

[0172] The variables in the above model are It is a nonlinear optimization model with linear objectives and quadratic constraints to obtain the optimal control setpoints for reactive power dispatch of inverters.

[0173] The present invention also provides a voltage-reactive power dual-layer optimization device for coordinating a CVR regulating device and an inverter, comprising:

[0174] Three-phase power flow model establishment module, used to establish a three-phase power flow model based on branch power flow;

[0175] An approximate three-phase power flow model and a linearized three-phase power flow model establishing module, configured to establish an approximate three-phase power flow model and a linearized three-phase power flow model based on branch power flow according to the three-phase power flow model based on branch power flow;

[0176] Model building modules for modeling voltage regulators, capacitor banks, inverters, and voltage-sensitive loads separately;

[0177] The voltage-reactive power dual-layer optimization module is used to use a linearized three-phase power flow model to determine the setting values of traditional CVR regulation equipment and inverters to establish an upper-layer MILP model, and to correct the inverter setting values according to the approximate nonlinear optimal power flow model to obtain an approximate optimal feasible solution and a lower-layer NLP model; the upper-layer MILP model and the lower-layer NLP model are solved, and the voltage-reactive power dual-layer optimization is performed based on the results.

[0178] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0179] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0180] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0181] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0182] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are protected by the present invention.

Claims

1. A voltage-reactive power dual-layer optimization method for coordinating CVR regulation equipment and inverter, characterized in that: The following steps are involved: The first step is to establish a three-phase power flow model based on branch power flow; The three-phase power flow model based on branch power flow is: Rank-1 positive semidefinite (PSD) matrix Where, is the complex power of phase p, is the p-phase complex voltage, is the p-phase current, Φ i represents the phase set of busbar i, z ij is the phase sequence impedance matrix, G = (N, ε), where N represents the busbar set, ε represents the branch set, each branch (i, j) represents the connection form between two adjacent nodes i and j, {a, b, c} represents the three phases of the system, for each busbar; For p∈Φ i Phase, complex power demanded by the load at bus i, and are active and reactive power demands respectively; (pq)∈Φ ij The complex power of branch (i, j) is and Corresponding to active power and reactive power respectively; is the complex impedance matrix z of branch (i, j) ij An element of , where (pq): p∈Φ i ,q∈Φ j ; In the above formula, the first formula represents the voltage drop equation, the second formula represents the power balance equation, the third formula represents the power equation calculation formula, and the fourth formula is the rank 1 constraint; The second step is to establish an approximate three-phase power flow model and a linearized three-phase power flow model based on the branch power flow according to the three-phase power flow model of the branch power flow; The third step is to model the voltage regulator, capacitor bank, inverter and voltage-sensitive load respectively; In the fourth step, a linearized three-phase power flow model is used to determine the setting values of the traditional CVR regulation equipment and the inverter to establish an upper-level MILP model, and the setting values of the inverter are corrected according to the approximate nonlinear optimal power flow model to obtain an approximate optimal feasible solution and a lower-level NLP model; the upper-level MILP model and the lower-level NLP model are solved, and voltage-reactive power double-level optimization is performed based on the results.

2. The voltage-reactive power dual-layer optimization method for coordinating CVR regulation equipment and inverter according to claim 1, characterized in that: The approximate three-phase power flow model based on branch power flow is: Among them, for all (i, j)∈ε, the first and second formulas represent pp∈Φ ij The active power and reactive power flow equations of branch (i, j) are: and Assumed to be a constant, the first and second equations are and is linear; the third formula is the voltage drop equation between the two nodes of the (i, j) branch corresponding to p, In the equation, the third equation is linear; the fourth equation relates the complex power flow of each phase in branch (i, j) to the phase voltage and phase current, and the fourth equation is a nonlinear quadratic equality constraint; where p∈Φ i ; where (pq)∈Φ ij ; is the branch current and The phase difference between The phase voltages differ by approximately 120°, which satisfies: is a constant, S ij The elements of are approximated by diagonal moments; The phase currents of phases p and q are and Given busbar i, The phase difference of the upper phase current is approximately a constant variable. By solving the equivalent distribution system power flow when the load is a constant impedance model, the branch (i, j) is obtained. approximation; The linearized three-phase power flow model is:

3. The voltage-reactive power dual-layer optimization method for coordinating CVR regulation equipment and inverter according to claim 2, characterized in that: The model of the voltage regulator is: and a p is the turns ratio of the voltage regulator connected to the (i, j) branch p, that is, a p The value of is between 0.9 and 1.

1. Each step produces a change of 0.00625 pu, introducing an additional node i′. For each step position of the voltage regulator, i∈(1, 2, ..., 32), For a 0-1 variable, define a vector bi∈{0.9, 0.90625, ..., 1.1}, then and Where p∈Φ i ∩Φ j as follows: and in, and To express the above as and function, take the square of the above formula, and define and and 4. The voltage-reactive power dual-layer optimization method for coordinating CVR regulation equipment and inverter according to claim 3, characterized in that: The model of the capacitor is: in, Indicates the on / off status of the capacitor bank, is the rated single-phase reactive power of the capacitor bank, is the square of the voltage of busbar i of phase p, is the reactive power generated by the capacitor bank.

5. The voltage-reactive power dual-layer optimization method for coordinating CVR regulation equipment and inverter according to claim 4, characterized in that: The inverter model is: is the available reactive power of the inverter. The known active power generation of DG is equal to the predicted value. The reactive power output of DG depends on the rated value of the inverter. To connect i th The rated single-phase apparent power capacity of the inverter at DG, To predict active power generation.

6. The voltage-reactive power dual-layer optimization method for coordinating CVR regulation equipment and inverter according to claim 5, characterized in that: The specific method for modeling voltage-sensitive loads is as follows: First, the mathematical model of the ZIP model of the load connected to the p-phase of bus i is established as follows: Among them, k p，1 +k p，2 +k p，3 =1,k q，1 +k q，2 +k q，3 =1, and is the active power demand and reactive power demand of single-phase load under standard voltage; Then, the ZIP load model is transformed into an equivalent load model using the CVR coefficient. The CVR coefficient is defined as the ratio of the percentage of active or reactive power reduction to the percentage of bus voltage reduction. The CVR coefficient of active and reactive power reduction is CVR. p 、CVR q They are: in, and therefore Let V i p ≈V0 and It turns out that: The above formula is based on the load model of CVR coefficient A linear load model based on the CVR coefficient Embedded into the three-phase power flow model established in the second step; CVR coefficient CVR p and CVR q It can be estimated by the ZIP coefficient of the load, differentiating the ZIP model and assuming V0 = 1 p.u., to obtain: Assume V i p ≈V0, the CVR coefficient of the load is estimated by the ZIP coefficient: CVR p =2k p，1 +k p，2 、CVR q =2k q，1 +k q，2 。 7. The voltage-reactive power dual-layer optimization method for coordinating CVR regulation equipment and inverter according to claim 6, characterized in that: The specific method for establishing the upper-layer MILP model by using the linearized three-phase power flow model to determine the setting values of the traditional CVR regulation equipment and inverter is as follows: By controlling the voltage regulator, capacitor bank and inverter to minimize the system loss, the optimization model is as follows to minimize the sum of the three-phase active power injected into the substation bus: Objective function: The constraints are as follows: Linear AC power flow constraints: CVR-based load model: Voltage regulator model: Capacitor Bank Model: Inverter model: Voltage Constraints: The variables in the above model are It is a mixed integer programming model.

8. The voltage-reactive power dual-layer optimization method for coordinating CVR regulation equipment and inverter according to claim 7, characterized in that: The specific method of correcting the inverter setting value according to the approximate nonlinear optimal power flow model to obtain the approximate optimal feasible solution and the lower layer NLP model is as follows: Adjust the setting value of the inverter control variable to obtain the optimal feasible three-phase nonlinear power flow solution, assuming that the discrete control variable For fixed, a real-time correction model of the inverter setting value based on the approximate nonlinear optimal power flow model is established as follows: Objective function: The constraints are as follows: The constraints of the approximate nonlinear AC power flow equation defined at time t are: CVR-based load model: Voltage regulator model: Capacitor Bank Model: Inverter model: Voltage Constraints: The variables in the above model are The optimal control setpoints for inverter reactive power dispatch are obtained for a nonlinear optimization model with linear objectives and quadratic constraints.

9. A voltage-reactive dual-layer optimization device for coordinating CVR regulation equipment and inverter, characterized in that: include: Three-phase power flow model establishment module, used to establish a three-phase power flow model based on branch power flow; An approximate three-phase power flow model and a linearized three-phase power flow model establishing module, configured to establish an approximate three-phase power flow model and a linearized three-phase power flow model based on branch power flow according to the three-phase power flow model based on branch power flow; Model building modules for modeling voltage regulators, capacitor banks, inverters, and voltage-sensitive loads separately; A voltage-reactive power dual-layer optimization module is configured to use a linearized three-phase power flow model to determine the setting values of a traditional CVR regulator and an inverter to establish an upper-layer MILP model, and to modify the inverter setting values according to the approximate nonlinear optimal power flow model to obtain a near-optimal feasible solution and a lower-layer NLP model. The upper-layer MILP model and the lower-layer NLP model are then solved, and voltage-reactive power dual-layer optimization is performed based on the results. The three-phase power flow model based on branch power flow is: Rank-1 positive semidefinite (PSD) matrix Where, is the complex power of phase p, is the p-phase complex voltage, is the p-phase current, Φ i represents the phase set of busbar i, z ij is the phase sequence impedance matrix, G = (N, ε), where N represents the busbar set, ε represents the branch set, each branch (i, j) represents the connection form between two adjacent nodes i and j, {a, b, c} represents the three phases of the system, for each busbar; For p∈Φ i Phase, complex power demanded by the load at bus i, and are active and reactive power demands respectively; (pq)∈Φ ij The complex power of branch (i, j) is and Corresponding to active power and reactive power respectively; is the complex impedance matrix z of branch (i, j) ij An element of , where (pq): p∈Φ i ,q∈Φ j ; In the above formula, the first formula represents the voltage drop equation, the second formula represents the power balance equation, the third formula represents the power equation calculation formula, and the fourth formula is the rank 1 constraint.

Images