A centralized-distributed distribution network voltage control method

Abstract

A centralized - distributed distribution network voltage control method includes the following steps: Step 1: Construct a centralized - distributed voltage control model; Step 2: Construct the objective function of the centralized - distributed voltage control model; Step 3: Construct the constraint conditions of the objective function of the centralized - distributed voltage control model; Step 4: Solve the optimal solution of the objective function of the centralized - distributed voltage control model through an optimization algorithm. It can not only adjust and control the global voltage of the distribution network substation area, but also solve the problem of voltage fluctuations caused by the randomness of distributed photovoltaic power, so as to ensure that the global voltage of the distribution substation area does not exceed the limit and guarantee the safe and stable operation of the power grid.

Description

Technical Field

[0001] The present invention relates to a distribution network voltage control method and belongs to the field of distribution network operation and analysis. Background Art

[0002] The high proportion of photovoltaic grid connection has a significant impact on the low-voltage distribution network, manifested in: 1) The power flow changes from unidirectional to bidirectional, and obvious reverse power flow is likely to occur especially under strong light, which may lead to voltage exceeding the upper limit; 2) The uncertainty of network operation increases significantly, and the node voltage fluctuates frequently and rapidly. Therefore, how to effectively weaken the above adverse effects and ensure the economic and safe operation of the network has become the focus of current research.

[0003] There are some deficiencies in simply using centralized or distributed control modes to regulate the reactive power of photovoltaic inverters. Although centralized control can effectively coordinate the reactive power output of different photovoltaic inverters, the data transmission delay and large computational amount restrict the response of centralized control to the rapid change of photovoltaic power. At the same time, the prediction error will also significantly affect the effect of centralized control. Although distributed control can quickly respond to the rapid and random fluctuations of photovoltaic power, the lack of information interaction makes it difficult to coordinate equipment. The access position of traditional voltage regulating equipment is relatively fixed, the regulation frequency is not high, and the amount of data to be transmitted is not large, so the centralized control mode is suitable; while the number of distributed photovoltaics is large and the positions are scattered, so the distributed control mode is suitable. Therefore, combining the two control modes can not only control the global voltage of the distribution network substation area, but also quickly respond to the rapid change of distributed photovoltaic power. Summary of the Invention

[0004] The purpose of the present invention is to solve the shortcomings that the current centralized control method cannot quickly respond to photovoltaic prediction errors and power fluctuations, and the problem that it is difficult to effectively coordinate between devices in distributed control, and thus propose a centralized-distributed distribution network voltage control method, which can not only adjust and control the global voltage of the distribution network substation area, but also solve the voltage fluctuation problem caused by the randomness of distributed photovoltaic power, so as to ensure that the global voltage of the distribution substation area does not exceed the limit and guarantee the safe and stable operation of the power grid.

[0005] A centralized-distributed distribution network voltage control method includes the following steps:

[0006] Step 1: Construct a centralized-distributed voltage control model;

[0007] Step 2: Construct the objective function of the centralized-distributed voltage control model;

[0008] Step 3: Construct the constraint conditions of the objective function of the centralized-distributed voltage control model;

[0009] Step 4: Solve the optimal solution of the objective function of the centralized-distributed voltage control model through an optimization algorithm.

[0010] Step 1 is specifically as follows:

[0011] Step 1.1: Establish a centralized control model

[0012] The control devices in the centralized control part are OLTC and CB. Voltage control models for OLTC and CB need to be established respectively;

[0013] Step 1.2: Establish a distributed control model

[0014] The control device in the distributed control part is a photovoltaic inverter. A voltage control model for the photovoltaic inverter needs to be established.

[0015] The voltage control strategy of the OLTC voltage control model is target - dead zone - time delay control, that is, when the difference ΔU between the measured voltage and the set voltage exceeds the threshold U BW , and when the holding time ΔT of the voltage difference i exceeds the set threshold time ΔT set , OLTC starts to adjust the gear position, expressed as the following formula:

[0016]

[0017] In the formula, is the gear position number of OLTC at time t; and are the minimum and maximum values of the OLTC gear position respectively; is the adjustable times of OLTC at node i; is the maximum adjustable times of OLTC; is the gear position number of OLTC at time t + Δt.

[0018] The CB voltage control model is expressed as:

[0019]

[0020] In the formula, is the reactive power output by CB; is the reactive power of a single group of CB; is the number of operating groups of CB; and are the minimum and maximum values of the reactive power output by CB; is the maximum number of operating groups that CB can operate.

[0021] Step 1.2 is specifically as follows:

[0022] The relationship between the reactive power and active power output by the photovoltaic inverter can be expressed as:

[0023]

[0024] Where: Q PV is the reactive power output by the PV inverter; S PV is the apparent power of the PV inverter; P PV is the active power of the PV inverter;

[0025] Under the voltage-reactive power droop control mode with a fixed coefficient, the target value of the reactive power of the PV inverter is calculated according to the following formula:

[0026] Q obj = Q 0 + K qv (U ref - U meas )U meas <U L or U meas >U H ;

[0027] Where: Q obj is the target value of the reactive power of the distributed PV inverter, in per-unit value (pu); Q 0 is the reactive power corresponding to the grid connection point voltage U meas of the distributed PV inverter, in per-unit value (pu); K qv is the reactive voltage droop control coefficient; U ref is the reference value of the grid connection point voltage (the rated voltage can be taken), in per-unit value (pu); U H is the upper limit of the dead zone voltage; U L is the lower limit of the dead zone voltage;

[0028] Under the voltage-reactive power droop control mode with a variable coefficient, the magnitude of the droop coefficient should be proportional to its reactive power capacity:

[0029]

[0030] Where: is the maximum reactive power that the i-th PV inverter can provide at time t; n is the number of PV inverters in the area; α is set according to the reactive power demand;

[0031] The constraint condition of the day-ahead PV inverter voltage control model is the reactive power output of the PV inverter, that is:

[0032] Q PV,min ≤ Q PV ≤ Q PV,max ;

[0033] Where, Q PV is the reactive power output by the PV inverter; Q PV,min is the lower limit value of the reactive power of the PV inverter; Q PV,maxIt is the upper limit of the reactive power of the PV inverter.

[0034] Step 2 is specifically as follows:

[0035] Taking the minimum of the distribution network power loss P Loss and the minimum of the voltage deviation value U DV of each node as the day-ahead control objectives:

[0036]

[0037] In the formula, where i and j represent the starting and ending nodes of the line, N represents the number of system nodes, U i represents the voltage of the starting node of the line, U j represents the voltage of the ending node of the line; cosθ ij represents the cosine of the node voltage phase angle, g ij represents the line conductance; U N represents the rated value of the node voltage; U max represents the maximum of the node voltage, U min represents the minimum of the node voltage;

[0038] Weigh the above two objective functions respectively and form the objective function f of the centralized-distributed voltage control model:

[0039] minf = α 1 P Loss + α 2 U DV ;

[0040] α 1 + α 2 = 1;

[0041] In the formula, α 1 and α 2 are weight coefficients, which can be selected according to the requirements of distribution network optimization.

[0042] The constraint conditions in Step 3 are:

[0043] AC power flow equation constraint

[0044]

[0045] In the formula: P Gi and Q Gi are the active power and reactive power injected at node i respectively; P Li and Q Li are the active power and reactive power of the load at node i respectively; δ ij is the phase difference between the voltage vectors of node i and node j; G ij and B ijThey are the conductance and susceptance of the line between node i and node j respectively; n is the total number of nodes in the power grid.

[0046] Voltage and current scenario constraints

[0047] U i,min ≤U i ≤U i,max ;

[0048] I i,min ≤I i ≤I i,max ;

[0049] In the formula: U i,max and U i,min represent the upper and lower limit values of the voltage of grid node i respectively; I i,min and I i,max represent the upper and lower limit values of the current of grid node i respectively; U i and I i represent the voltage and current of grid node i respectively.

[0050] Step 4 is specifically as follows:

[0051] 41) Set the population size of the particle swarm of the centralized-distributed voltage control model. The position information of the particle swarm consists of power grid parameters, distributed photovoltaic parameters, OLTC and CB parameters, and load parameters. Initialize the particle position and velocity.

[0052] 42) Calculate the voltage deviation of the distribution network nodes and the network loss as the fitness value of the particles. Evaluate the fitness values of all particles, and save the global optimal position and individual extreme points of the particles respectively.

[0053] 43) Judge the maximum number of iterations or the convergence accuracy. If the network loss P Loss of the distribution network is minimized and the voltage deviation values U DV of each node are minimized, then end; if the requirements are not met yet, update the position and velocity information of the particles and repeat steps 42) and 43).

[0054] The control objects of the centralized control mode of the present invention mainly include on-load tap-changing transformers (OLTCs) and intelligent capacitors (CBs), which are controlled once every 6 hours; the control object of the distributed control mode is the photovoltaic inverter, and the inverter is controlled in real time. The control objectives of the centralized-distributed voltage control method are to minimize the photovoltaic active power curtailment, the distribution network loss, and the voltage deviation of each node.

[0055] Compared with the existing centralized control and distributed control modes, the present invention can not only adjust and control the global voltage of the distribution network substation in a controllable manner, but also solve the problem of voltage fluctuations caused by the randomness of distributed photovoltaic power, so as to ensure that the global voltage of the distribution substation does not exceed the limit and guarantee the safe and stable operation of the power grid. Detailed implementation manners

[0056] To make the implementation and advantages of the technical method of the present invention clear and understandable, the implementation process is described in detail as follows:

[0057] Step 1: Construct a centralized-distributed voltage control model

[0058] The control devices in the centralized control part of the present invention are OLTC and CB. Therefore, it is necessary to establish the control models of OLTC and CB.

[0059] 1) OLTC model

[0060] The voltage control strategy of the OLTC proposed by the present invention is target-dead zone-delay control, that is, when the difference ΔU between the measured voltage and the set voltage exceeds the threshold U BW When the voltage difference holding time ΔT i exceeds the set threshold time ΔT set , the OLTC will start to adjust the gear. The delay setting is to ensure that the voltage deviation is not instantaneous but a continuous process.

[0061]

[0062] In the formula, is the number of gears of the OLTC at time t; and are the minimum and maximum values of the OLTC gear respectively; is the adjustable times of the OLTC at node i; is the maximum adjustable times of the OLTC; is the number of gears of the OLTC at time t + Δt.

[0063] 2) Capacitor (CB) model

[0064]

[0065] In the formula, is the reactive power output by the CB; is the reactive power of a single group of the CB; is the number of groups of the CB put into operation; and are the minimum and maximum values of the reactive power output by the CB respectively; is the maximum number of groups that the CB can be put into operation.

[0066] In the distributed control model of the present invention, the control device in the part is a photovoltaic inverter, and the relationship between the reactive power output and the active power of the photovoltaic inverter can be expressed as:

[0067]

[0068] In the formula: Q PV is the reactive power output by the photovoltaic inverter; S PV is the apparent power of the photovoltaic inverter; P PV is the active power of the photovoltaic inverter.

[0069] Under the traditional fixed - coefficient voltage - reactive power (U - Q) droop control (Voltage - VAR Droop Control, VVDC) mode, the target value of the reactive power of the photovoltaic power station is calculated according to the following formula:

[0070] Q obj = Q 0 + K qv (U ref - U meas ) U meas <U L or U meas >U H (4)

[0071] In the formula: Q obj is the target value of the distributed photovoltaic reactive power, in per - unit value (pu); Q 0 is the power corresponding to the distributed photovoltaic connection point voltage of U meas , in per - unit value (pu); K qv is the reactive voltage droop control coefficient; U ref is the reference value of the connection point voltage (the rated voltage can be taken), in per - unit value (pu); U H is the upper limit of the dead - zone voltage; U L is the lower limit of the dead - zone voltage.

[0072] Under the variable - coefficient VVDC mode, the magnitude of the droop coefficient should be proportional to its reactive power capacity.

[0073]

[0074] In the formula: is the maximum reactive power that the i - th photovoltaic inverter can provide at time t; n is the number of photovoltaic inverters in the area; α is adjusted according to the reactive power demand.

[0075] The constraint condition of the voltage control model of the photovoltaic inverter for the day - ahead is the reactive power output of the photovoltaic inverter, that is:

[0076] Q PV,min ≤QPV ≤Q PV,max (6)

[0077] Step 2: Construct the objective function of the centralized-distributed voltage control model

[0078] The day-ahead control objective is the network loss P of the distribution network Loss minimum and the voltage deviation value U of each node DV minimum

[0079]

[0080]

[0081] In the formula, where i and j represent the start and end nodes of the line, N represents the number of system nodes, and U i , U j represent the start and end voltages of the line; cosθ ij represents the cosine of the node voltage phase angle, and g ij represents the line conductance; U N represents the rated value of the node voltage; U max , U min represent the maximum and minimum values of the node voltage.

[0082] For the convenience of subsequent solution and analysis, the above three objective functions are weighted respectively, and the objective function f of the centralized-distributed voltage control model is formed:

[0083] minf = α 1 P Loss + α 2 U DV (9)

[0084] α 1 + α 2 = 1 (10)

[0085] In the formula, α 1 and α 2 are weight coefficients, which can be selected according to the requirements of distribution network optimization.

[0086] Step 3: Determine the constraint conditions of the centralized-distributed voltage control model

[0087] 1) AC power flow equation constraint

[0088]

[0089] In the formula: P Gi and Q Gi are the active power and reactive power injected at node i respectively; P Li and Q LiThey are the active power and reactive power of the load at node i; δ ij is the phase difference between the voltage vectors of node i and node j; G ij and B ij are the conductance and susceptance of the line between node i and node j respectively; n is the total number of nodes in the power grid.

[0090] 2) Voltage and current scenario constraints

[0091] U i,min ≤U i ≤U i,max (12)

[0092] I i,min ≤I i ≤I i,max (13)

[0093] In the formula: U i,max and U i,min represent the upper and lower limit values of the voltage at node i of the power grid respectively; I i,min and I i,max represent the upper and lower limit values of the current at node i of the power grid respectively.

[0094] Step 4: Use the optimization algorithm to solve the optimal solution of the objective function of the centralized - distributed voltage control model

[0095] 41) Set the population size of the particle swarm of the centralized - distributed voltage control model. The position information of the particle swarm consists of power grid parameters, distributed PV parameters, OLTC and CB parameters, and load parameters. Initialize the particle position and velocity.

[0096] 42) Calculate the voltage deviation of the distribution network nodes and the network loss as the fitness value of the particles. Evaluate the fitness values of all particles, and save the global optimal position and individual extreme points of the particles respectively.

[0097] 43) Judge the maximum number of iterations or the convergence accuracy. If the formulas (7) and (8) are satisfied, then end; if the requirements are not met, update the position and velocity information of the particles and repeat steps 42) and 43).

Claims

1. A centralized-distributed distribution network voltage control method, characterized in that: The following steps are involved: Step 1: Construct a centralized-distributed voltage control model; Step 2: Construct the objective function of the centralized-distributed voltage control model; Step 3: Construct the constraints of the objective function of the centralized-distributed voltage control model; Step 4: Obtain the optimal solution of the objective function of the centralized-distributed voltage control model through an optimization algorithm; Step 1 is as follows: Step 1.1: Establish a centralized control model The control equipment of the centralized control part is OLTC and CB, and the voltage control models of OLTC and CB need to be established respectively; Step 1.2: Build a distributed control model The control device in the distributed control part is a photovoltaic inverter, and a photovoltaic inverter voltage control model needs to be established; Step 2 is as follows: The distribution network loss P Loss Minimum and each node voltage deviation value U DV The minimum is the day-ahead control target: In the formula, i and j represent the starting and ending nodes of the line, N represents the number of system nodes, and U i Indicates the line start node voltage, U j Represents the voltage at the end node of the line; cosθ ij represents the node voltage phase angle cosine, g ij Indicates line conductance; U N Indicates the node voltage rating; U max Indicates the maximum node voltage, U min Indicates the minimum value of node voltage; The above two objective functions are weighted respectively to form the objective function f of the centralized-distributed voltage control model: minf=α1P Loss +α2U DV ; α1+α2＝1； In the formula, α1 and α2 are weight coefficients, which can be selected according to the needs of distribution network optimization.

2. A centralized-distributed distribution network voltage control method according to claim 1, characterized in that: The voltage control strategy of the OLTC voltage control model is target-dead zone-delay control, that is, when the difference ΔU between the measured voltage and the set voltage exceeds the threshold U BW , and when the voltage difference is maintained for time ΔT i Exceeds the set threshold time ΔT set When , OLTC starts to adjust the gear, which is expressed as follows: In the formula, is the number of gears of OLTC at time t; and They are the minimum and maximum values ​​of the OLTC gear respectively; is the adjustable number of OLTC at node i; The maximum number of times that OLTC can be adjusted; is the number of gears of OLTC at time t+Δt.

3. A centralized-distributed distribution network voltage control method according to claim 1 or 2, characterized in that: The CB voltage control model is expressed as: In the formula, is the reactive power output by CB; is the reactive power of a single CB group; is the number of CB groups put into operation; and is the minimum and maximum value of reactive power output by CB; The maximum number of CB groups that can be put into operation.

4. A centralized-distributed distribution network voltage control method according to claim 1, characterized in that: Step 1.2 is as follows: The relationship between the reactive power and active power output of the photovoltaic inverter can be expressed as: Where: Q PV is the reactive power output by the photovoltaic inverter; S PV is the apparent power of the photovoltaic inverter; P PV is the active power of the photovoltaic inverter; In the constant coefficient voltage-reactive power droop control mode, the reactive power target value of the photovoltaic inverter is calculated according to the following formula: Q obj = Q0 + K qv (U ref - U meas )U meas <U L or U meas >U H ; Where: Q obj is the reactive power target value of the distributed photovoltaic inverter, in per unit value (pu); Q0 is the grid-connected point voltage of the distributed photovoltaic inverter, U meas The reactive power corresponding to the time is expressed in per unit value (pu); K qv is the reactive voltage droop control coefficient; U ref is the grid connection point voltage reference value, in per unit value (pu); U H is the upper limit of the dead zone voltage; U L is the lower limit of the dead zone voltage; In the variable coefficient voltage-reactive power droop control mode, the droop coefficient should be proportional to its reactive power capacity: Where: is the maximum reactive power that the i-th PV inverter can provide at time t; n is the number of PV inverters in the region; α is set according to the reactive power demand; The constraint condition of the current PV inverter voltage control model is the reactive power output of the PV inverter, namely: Q PV,min ≤Q PV ≤Q PV,max ; In the formula, Q PV is the reactive power output by the photovoltaic inverter; Q PV,min is the lower limit of reactive power of the photovoltaic inverter; Q PV,max It is the upper limit of reactive power of PV inverter.

5. A centralized-distributed distribution network voltage control method according to claim 1, characterized in that: The constraints in step 3 are: AC Power Flow Equation Constraints Where: P Gi and Q Gi are the active power and reactive power injected by node i respectively; P Li and Q Li are the active power and reactive power of the load at node i respectively; δ ij is the phase difference between the voltage vectors at node i and node j; G ij and B ij are the conductance and susceptance of the line between node i and node j respectively; n is the total number of nodes in the power grid; Voltage and current scenario constraints IN i,min ≤U i ≤U i,max ; I i,min ≤I i ≤I i,max ; Where: U i,max , U i,min Respectively represent the upper and lower limits of the voltage of the grid node i; I i,min ,I i,max Respectively represent the upper and lower limits of the current of the grid node i; U i ,I i They represent the voltage and current of grid node i respectively.

6. A centralized-distributed distribution network voltage control method according to claim 1, characterized in that: Step 4 is as follows: 41) The population size of the particle swarm of the centralized-distributed voltage control model is set. The position information of the particle swarm is composed of grid parameters, distributed photovoltaic parameters, OLTC and CB parameters, and load parameters. The particle position and speed are initialized; 42) Calculate the voltage deviation of the distribution network node and the network loss as the fitness value of the particle, evaluate the fitness values ​​of all particles, select the global optimal position of the particle and the individual extreme point and save them separately; 43) Determine the maximum number of iterations or convergence accuracy. If the distribution network loss P is satisfied Loss Minimum and each node voltage deviation value U DV If the requirement is not met, the particle position and velocity information is updated and steps 42) and 43) are repeated.

Images