KR20210133062A - Optimal setting method of volt-var curve - Google Patents

Abstract

According to the present invention, a fitness evaluating method about a voltage-reactive power curve comprises the following steps of: selecting an initial voltage-reactive power curve; configuring a multi-objective function for at least two of voltage deviation, loss, and peak of contributing reactive power for setting a parameter set of the voltage-reactive power curve; and determining an optimal voltage-reactive power curve having an optimal parameter set from a viewpoint selected from the voltage deviation, the loss, and the peak of the contributing reactive power by using the multi-objective function.

Description

Evaluation method of fitness for voltage-reactive power curve {OPTIMAL SETTING METHOD OF VOLT-VAR CURVE}

The present invention relates to a method for optimally setting a voltage-reactive power curve for controlling an inverter, etc., and more particularly, to a method for evaluating the suitability of a voltage-reactive power curve for improving the performance of a smart inverter.

Voltage-reactive power (Volt-Var) function of smart inverters is in the spotlight as a way to solve voltage instability problems such as overvoltage due to the increase in distributed power connection.

In order to effectively control each distributed power source in a power distribution system or microgrid, data such as voltage and current are collected and voltage-reactive power (Volt-Var), frequency-active power, etc. are displayed as control curves.

1 is a graph showing a curve of a general Volt-Var function of a smart inverter according to the prior art.

The voltage-reactive power (Volt-Var) function of the smart inverter sets a command value of reactive power according to the Point of Common Coupling (PCC) voltage as shown in FIG. 1 .

A curve is set according to the parameters of the Volt-Var function, and the result affects the voltage deviation and loss corresponding to the performance index in the distribution system.

In addition, if the amount of reactive power contributing to the distribution system increases by capacity in the Power Conditioning System (PCS) of distributed power sources such as solar power generation, it may have the same effect as the decrease in active power of distributed power generation.

Republic of Korea Registration No. 10-1043572

The present invention intends to propose a suitability evaluation method for the voltage-reactive power curve as an optimal parameter setting method (algorithm) of the Volt-Var function of a smart inverter to improve performance during actual operation of the system.

In the present invention, when setting the parameters of the Volt-Var function of the smart inverter, considering the contribution of voltage deviation, loss, and reactive power, which are performance indicators of the system, an optimal parameter setting method that can improve the necessary performance indicators (algorithm) is proposed.

According to an aspect of the present invention, a method for evaluating fitness for a voltage-reactive power curve includes: selecting an initial voltage-reactive power curve; Constructing a multi-objective function for at least two or more of voltage deviation, loss, and peak of contributing reactive power for setting a parameter set of a voltage-reactive power curve; and using the multi-objective function to determine an optimal voltage-reactive power curve having an optimal parameter set in a selected viewpoint among voltage deviations, losses, and peaks of contributing reactive power.

Here, in the step of selecting the initial voltage-reactive power curve, a Moderate curve of the voltage-reactive power function may be selected as the initial voltage-reactive power curve.

Here, in the step of determining the optimal voltage-reactive power curve, a process of setting a specific parameter set, and calculating the fitness of the multiple objective function set with the specific parameter set with multiple simulation data for a predetermined test period an iterative step of repeatedly performing the process for each parameter set that is a candidate target, and selecting the one with the highest degree of fit; and determining an optimal voltage-reactive power curve with the selected specific parameter set.

Here, in the step of determining the optimal voltage-reactive power curve, a particle swarm optimization (PSO) technique using the parameter set as a particle may be applied.

Here, the voltage-reactive power curve may follow the following equation.

Here, the multi-objective function may follow the following equation.

는 시간(t)에서의 버스(i)의 전압 편차 지수, P_t(vv)는 volt-var 함수의 파라미터에 따른 시간(t)에서의 시스템 손실, Q_i(vv)는 volt-var 기능의 매개 변수에 따른 스마트 인버터의 무효전력 피크)(i is the bus with distributed generation, t is the time, vv _new is the parameter for the newly updated volt-var curve function for each individual test, and vv _con is the optimal tentative volt- in the initial or previous personal loop. var function parameters for the curve function, w is the weight of each evaluation index, V _i,t is the voltage of the bus (i) at time (t) according to the parameters of the volt-var function,

is the voltage deviation exponent of the bus (i) at time (t), P _t (vv) is the system loss at time (t) according to the parameter of the volt-var function, Q _i (vv) is the voltage deviation index of the volt-var function Reactive power peak of smart inverter according to parameters)

Here, the multi-objective function may reflect the constraint according to the following equation.

(N _const is the constraint setting, w _c is the penalty coefficient, PF _c is the weight for each constraint)

If the suitability evaluation method for the voltage-reactive power curve according to the spirit of the present invention of the above-described configuration is implemented, there is an advantage in that the parameters of the Volt-Var function of the smart inverter can be optimally set according to the actual operating environment.

The optimal setting method of the voltage-reactive power curve of the present invention has the advantage of minimizing the voltage deviation and system loss representing the indicator of the power system, and suppressing the peak of the reactive power in order not to be affected by the output of the distributed power. have.

1 is a graph showing a curve of a general Volt-Var function of a smart inverter according to the prior art.
2 is a flowchart illustrating an embodiment of a method for optimally setting a voltage-reactive power curve according to the spirit of the present invention;
Figure 3 is a graph showing a curve of the Volt-Var function according to the operation policy (condition) of the smart inverter according to the prior art.
4 is a flowchart illustrating a process of determining a personal best and a global best.
5 is a system configuration diagram showing a test model of the distribution system for verification of the proposed method.
6 is a graph showing the power usage pattern of solar power generation and load for one year in Korea.
7 is a graph showing a voltage pattern for one year simulated when the Volt-Var function (set to “Moderate”) of the existing smart inverter is applied.
8A is a case comparison graph for verifying the voltage indicator improvement effect.
8B is a graph showing the voltage index for each time period of the Volt-Var function curve for each case.
9A is a case comparison graph for verifying the system loss improvement effect.
9B is a graph showing the system loss over time of the Volt-Var function curve for each case.
10A is a case comparison graph for verifying the contributing reactive power peak suppression effect.
Figure 10b is a graph showing the reactive power peak amount for each time period of the Volt-Var function curve for each case.

In describing the present invention, terms such as first, second, etc. may be used to describe various components, but the components may not be limited by the terms. The terms are only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component.

When a component is referred to as being connected or connected to another component, it may be directly connected or connected to the other component, but it can be understood that other components may exist in between. .

The terms used herein are used only to describe specific embodiments, and are not intended to limit the present invention. The singular expression may include the plural expression unless the context clearly dictates otherwise.

In this specification, the terms include or include are intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, and includes one or more other features or numbers, It may be understood that the existence or addition of steps, operations, components, parts or combinations thereof is not precluded in advance.

In addition, shapes and sizes of elements in the drawings may be exaggerated for clearer description.

2 is a flowchart illustrating an exemplary embodiment of a method for optimally setting a voltage-reactive power curve (ie, a fitness evaluation method) according to the spirit of the present invention.

The illustrated voltage-reactive power curve optimal setting method includes the steps of selecting an initial voltage-reactive power curve (S120); voltage - for setting a parameter set of the reactive power curve, voltage deviation, loss, and configuring a multi-objective function for at least two or more peaks of contributing reactive power (S140); and determining an optimal voltage-reactive power curve having an optimal parameter set from a viewpoint selected among voltage deviation, loss, and peak of contributing reactive power using the multi-objective function (S200).

3 is a graph showing a curve of the Volt-Var function according to the operation policy (condition) of the smart inverter according to the prior art.

Figure 3 shows the standard default setting of the curve of the Volt-Var function. The standard setting can be expressed as "Aggressive", "Moderate", "Mild" depending on the slope. The standard volt-var function has a non-response band depending on the nominal voltage (except "aggressive"). Regions other than the refractory band can be configured linearly for absorption or injection of reactive power to help maintain the PCC voltage. The peak of reactive power for the curve shown is only affected by inverter capacity and driver priority.

As shown, even in the prior art, it is possible to adjust the Volt-Var curve to some extent according to the management (operation) policy of the power system. For example, in FIG. 3, Aggressive can be applied to inverters such as ESS, which has a policy or purpose to compensate reactive power of the system most quickly, Moderate can be applied to general inverters, and Mild is an inverter such as a generator where stable operation is important. can be applied to

In the present invention, in addition to adjustment according to the purpose or policy in terms of urgency vs stability for power management as shown in FIG. 3, it is specifically reflected in actual power management, and voltage deviation, system loss, and peak of reactive power indicating an indicator of the power system A method of deriving an optimal Volt-Var curve that can be directly improved is presented.

In the present invention, in order to derive the optimal Volt-Var curve, the optimal position for each reference point (which becomes a broken point) defining the shape of the Volt-Var curve is determined.

As each reference point of the Volt-Var curve, as shown in FIG. 2, the Aggressive curve has three reference points, and the Moderate and Mild curves have five reference points. When the origin is fixed or excluded, two and four are one by one. have reduced reference points. The set of reference points defining a specific curve may constitute one specific set of parameters.

In the present invention, particle swarm optimization (PSO) may be applied to derive the optimal positions of the reference points.

PSO is a computational method that optimizes a problem by iteratively improving a candidate solution by taking a motif from the movement of organisms in a flock or school of fish. have In addition to PSO, optimization methods such as genetic algorithm (GA) and Ant Colony algorithm for solving distribution system problems may be used.

In using the above-described optimization method such as PSO, first, an initial point of the optimization operation must be designated. For this, the step of selecting an initial voltage-reactive power curve (S120) is performed. For example, in the following, a Moderate curve is selected among the curves of FIG. 3 as the initial voltage-reactive power curve. In another implementation, an Aggressive curve or a Mild curve is selected according to the location and timing characteristics of the inverter. You can also choose

Equation 1 below is for defining a Moderate curve among the curves of FIG. 3, and represents the reactive power command value of the smart inverter according to the PCC voltage.

이하인 경우, 무효전력(

)을 기여하여 저전압 문제를 해결하고, 전압이

와

사이에 해당하는 경우, 선형적인 기울기를 그래프에 따라, 전압에 해당하는 무효전력을 기여한다. When the smart inverter is controlled according to the Volt-Var curve according to the above equation, the voltage is

If less than, reactive power (

) to solve the low voltage problem, and

Wow

If it corresponds to between, according to the linear slope graph, it contributes the reactive power corresponding to the voltage.

와

사이에 해당하는 경우, 불응 영역(Dead-Band)에 해당하므로 무효전력을 기여하지 않는다. On the other hand, the voltage

Wow

If it falls between, it does not contribute reactive power because it corresponds to a dead-band.

와

사이에 해당하는 경우, 선형적인 기울기를 그래프에 따라, 전압에 해당하는 무효전력을 기여하고, 전압이

이상인 경우, 무효전력(

)을 기여하여 과전압 문제를 해결한다.On the other hand, the voltage

Wow

If it falls between, according to the linear slope graph, it contributes the reactive power corresponding to the voltage, and the voltage is

In case of abnormality, reactive power (

) to solve the overvoltage problem.

Next, the step of constructing the multi-objective function ( S140 ) will be described.

Equation 2 below illustrates a multi-objective function for minimizing the peak of voltage deviation, loss, and contributing reactive power when setting the parameters of the Volt-Var function of the smart inverter.

In the above equation, i is a Bus with distributed generation, and t is time.

In the above equation, vv _new is the parameter (Vvv1, Vvv2, Vvv3, Vvv4, Qvv1, Qvv4) for the newly updated volt-var curve function for each individual test, and vv _con is the optimal value in the initial or previous personal loop. Function parameters for the temporarily selected volt-var curve function. w is the weight (weight) of each evaluation index (voltage deviation, line loss, peak of reactive power).

는 시간(t)에서의 버스(i)의 전압 편차 지수를 나타낸다. 이때, 레퍼런스 전압 V_ref는 예컨대, 22.9[kV]로 설정할 수 있다.In the above formula, V _i,t represents the voltage of the bus (i) at time (t) according to the parameters of the volt-var function,

denotes the exponent of the voltage deviation of the bus i at time t. In this case, the reference voltage V _ref may be set to, for example, 22.9 [kV].

In the above equation, P _t (vv) is the system loss at time (t) according to the parameter of the volt-var function, and Q _i (vv) is the reactive power peak of the smart inverter according to the parameter of the volt-var function .

The first item of Equation 2 represents a voltage deviation according to a parameter newly set compared to the initial curve method (“Moderate”), and the second item of Equation 2 is newly compared to the initial curve method (“Moderate”). It represents the loss of the system according to the set parameter, and the third item in Equation 2 represents the maximum value of the contribution of reactive power according to the newly set parameter compared to the initial curve method (“Moderate”).

ω indicates the weight of each item, and the higher the value, the more influence on the result.

Equation 2 above becomes an objective function to be optimized (minimized) in order to improve system performance in the environment for each site. This is because the volt-var function setting of the smart inverter affects the system loss and voltage deviation, which are indicators of system performance. System (line) losses are related to DSO and customer revenue, and voltage variations are directly related to power quality and system operation plan.

In addition, in some cases, if the volt-var function function is set, there is a possibility of a side effect of reducing the output of PV in relation to MPPT, etc. Therefore, since PV output is directly related to the interests of individual operators, it is difficult to install a control device that causes output reduction in smart inverters.

In consideration of the above circumstances, in the present invention, three peaks of voltage deviation, system loss, and reactive power are presented as a viewpoint (item or condition) to be considered when setting the parameters for determining the optimal Volt-Var function curve. .

However, when Equation 2 is applied as it is, there is a risk that the optimal curve is determined in a form that does not satisfy the basic condition of power system control. In order to block the above risk, in determining the optimal parameter, it is preferable to reflect the constraint for power system control.

Equation 3 below shows a fitness value for finding an optimal parameter in consideration of the above constraints.

In the above equation, N _const is a constraint setting, w _c is a penalty coefficient, and PF _c is a weight for each constraint.

Equation 3 represents a goodness-of-fit value for improving system performance in each simulation iteration period (eg, 1 year) using Equation 2 and the constraints. The particle with the minimum (optimal) goodness-of-fit value (ie the parameter, vv _new ) represents the parameter of the volt-var function that can maximize the system performance. In other words, when the constraint is violated due to the penalty coefficient according to Equation 3, it is not considered as an optimization parameter.

Equation 4 below represents the voltage maintenance range of the distribution system as the first constraint.

Equation 5 below represents the setting range of parameters of the smart inverter Volt-Var function as the second constraint.

By designating a range of parameters as in Equation 4 and/or Equation 5, an effect of improving system performance can be achieved while maintaining the essential shape of the Volt-Var function curve.

According to the PSO algorithm, etc., the step of determining the optimal voltage-reactive power curve (S200) includes a process of setting a specific parameter set, and a plurality of simulation data for a predetermined test period (1 year) with the specific parameter set an iterative step of repeatedly performing the process of calculating the fitness of the multi-objective function set to . and determining an optimal voltage-reactive power curve with the selected specific parameter set.

In the case of the PSO algorithm, candidate particles that are given as initial values or optimally determined in the previous process at a specific number of times (that is, the parameter set) are selected to select candidate particles that can be the optimal particle positions, and each candidate A fitness value for each candidate particle is calculated by applying the multi-objective function according to Equations 1 to 5 above with respect to the particles, and has a minimum or maximum fitness value according to conditions. A personal best is determined as an optimal value determined at each of the number of candidate particles.

The process of determining the personal best is repeated until a predetermined termination condition is satisfied to determine the final global best.

4 is a flowchart illustrating a process in which step S200 is performed in a manner of determining the above-described personal best and global best. That is, FIG. 4 shows a detailed flow using PSO as an algorithm for finding the optimal parameter for parameter optimization of the Volt-Var function of the smart inverter for system performance optimization.

In the illustrated step S220, in association with the step S120 of FIG. 2, particles may be initialized with reference points of the Volt-Var function curve selected in the step S120.

In the illustrated step S240, a fitness value according to Equation 3 is calculated for each particle (position).

In the illustrated step S260, the above-described personal best or global best is determined according to the loop being executed.

The illustrated step S280 is to iteratively calculate the optimal parameter of the multi-objective function (Equation 3) in consideration of the constraints by Equations 4 and/or 5, etc., until the termination condition is satisfied.

Hereinafter, effects on various constraints proposed by the present invention will be described.

5 shows a test model of the distribution system for verification of the proposed method. The illustrated test model consists of short-distance (Feeder #1), medium-distance (Feeder #2), and long-distance (Feeder #3) lines, and takes into consideration the characteristics of the Korean distribution system such as line type and transformer capacity.

6 is a one-year photovoltaic power generation and load power use pattern in Korea. The graph on the left shows the photovoltaic power generation pattern, and the graph on the right shows the power use pattern of the load. That is, FIG. 6 shows the pattern of solar power generation (irradiation amount) and power use for one year as test data for simulation verification of the proposed method.

The illustrated test pattern for power use may also be used in step S200 of FIG. 2 , that is, in the process of determining the personal best and global best of FIG. 4 .

As can be seen from the pattern shown, in Korea, depending on the surrounding environment such as irradiation amount and temperature, the solar power generation amount is highest in May, and the power usage pattern is the highest in summer and winter due to the cooling/heating load, and is highest during Lunar New Year and Chuseok. power consumption is drastically reduced.

7 shows a one-year voltage pattern simulated when the Volt-Var function (set to “Moderate”) of the existing smart inverter is applied. As shown, by applying the Volt-Var function of the smart inverter, it can be confirmed that the voltage deviation is reduced to some extent as the voltage pattern approaches 1 [p.u.].

In order to verify the effect of applying the optimized Volt-Var function according to the spirit of the present invention, first, cases will be classified and defined as shown in Table 1 below.

In the table above, W/O VV is a case in which the Volt-Var function of the smart inverter is not applied, Case A is a case in which the existing setting is made, and Case B is a case in which the weights of multiple objective functions are applied equally, Case C is a case in which more improvement effects can be expected in the voltage deviation factor by applying more weight to the voltage deviation, and in Case D, more improvement effects in the system loss factor can be expected by applying more weight to the system loss. Case E is a case in which more improvement effects can be expected in reactive power peak suppression by applying more weight to reactive power peak suppression.

8A and 8B are graphs illustrating a voltage index for each time period of a case comparison and a Volt-Var function curve for each case for verifying the improvement effect of the voltage index.

That is, in FIGS. 8A and 8B, comparative analysis is performed for each case for analyzing the voltage indicator improvement effect, and the voltage indicator improvement effect of the graph of FIG. 8B is most effectively improved in Case C, which has the highest weight (closer to 0, the nominal close to voltage).

On the other hand, in the graph of FIG. 8A , in Case C, which has the greatest effect of improving the voltage index, an optimal Volt-Var function curve without a deadband region is formed.

9A and 9B are graphs showing the system loss over time of case comparison and Volt-Var function curve for each case for verifying the system loss improvement effect.

That is, in FIGS. 9A and 9B, comparative analysis is performed for each case for analyzing the system loss improvement effect, and in the graph of FIG. 9B, it can be seen that Case D, which has the highest weight, improved the system loss improvement effect most effectively.

Meanwhile, the graph of FIG. 9B shows that the curve of the optimal Volt-Var function according to the system loss improvement effect is affected by various factors such as line impedance, distributed power and load power patterns, line length, and system configuration.

10a and 10b are graphs showing the reactive power peak amount for each time period of the case comparison and the Volt-Var function curve for each case for verifying the contributing reactive power peak suppression effect.

That is, FIGS. 10a and 10b perform a comparative analysis for each case for the analysis of the contributing reactive power peak suppression effect, and the graph of FIG. 10b shows that Case E with the highest weight for peak suppression of reactive power by time period is most effectively improved show that it has been

On the other hand, in the graph of FIG. 10A, it can be confirmed that the curve of the optimal Volt-Var function according to the suppression of the reactive power peak is formed as the curve in which Case E is the lowest in the voltage region where the peak occurs.

Table 2 below summarizes the simulation results described above.

It can be seen that Case C, in which the voltage deviation was given a high weight, minimized the sum of the maximum and minimum voltage values and the voltage index for one year.

Case D, in which the system loss was weighted high, improved the system loss by 6.2% compared to the case where the Volt-Var function was not applied, and 2% compared to the existing method. It can be seen that the suppression effect is obtained by about 75% compared to the existing method.

It can be seen that Case B, in which the same weight is applied to each system evaluation element, is the most improved in terms of evaluation index.

Those skilled in the art to which the present invention pertains should understand that the present invention may be embodied in other specific forms without changing the technical spirit or essential characteristics thereof, so the embodiments described above are illustrative in all respects and not restrictive. only do The scope of the present invention is indicated by the following claims rather than the detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention. .

S120: Step of selecting the initial voltage-reactive power curve
S140: Step of constructing a multi-objective function
S200: determining the optimal voltage-reactive power curve

Claims (7)

selecting an initial voltage-reactive power curve;
Constructing a multi-objective function for at least two or more of voltage deviation, loss, and peak of contributing reactive power for setting a parameter set of a voltage-reactive power curve; and
Determining an optimal voltage-reactive power curve having an optimal parameter set in a selected viewpoint among voltage deviation, loss, and peak of contributing reactive power using the multi-objective function
A fitness evaluation method for a voltage-reactive power curve including
According to claim 1,
In the step of selecting the initial voltage-reactive power curve,
A fitness evaluation method for a voltage-reactive power curve in which a Moderate curve of a voltage-reactive power function is selected as the initial voltage-reactive power curve.
According to claim 1,
In the step of determining the optimal voltage-reactive power curve,
the process of setting it to a specific set of parameters, and
Repeating the process of calculating the fitness of the multi-objective function set to the specific parameter set with multiple simulation data for a predetermined test period for each parameter set as a candidate target, and selecting the one with the highest degree of fitness ; and
Determining an optimal voltage-reactive power curve with a selected set of specific parameters
A fitness evaluation method for a voltage-reactive power curve including
4. The method of claim 3,
In the step of determining the optimal voltage-reactive power curve,
A fitness evaluation method for a voltage-reactive power curve by applying a particle swarm optimization (PSO) technique using the parameter set as a particle.
According to claim 1,
The voltage-reactive power curve is a voltage-reactive power curve according to the following equation.

6. The method of claim 5,
The multi-objective function is
A method for evaluating fitness for a voltage-reactive power curve according to the following equation.

(i is the bus with distributed generation, t is the time, vv _new is the parameter for the newly updated volt-var curve function for each individual test, and vv _con is the optimal tentative volt- in the initial or previous personal loop. var function parameters for the curve function, w is the weight of each evaluation index, V _i,t is the voltage of the bus (i) at time (t) according to the parameters of the volt-var function,

is the voltage deviation exponent of the bus (i) at time (t), P _t (vv) is the system loss at time (t) according to the parameter of the volt-var function, Q _i (vv) is the voltage deviation index of the volt-var function Reactive power peak of smart inverter according to parameters)
7. The method of claim 6,
The multi-objective function is
A fitness evaluation method for the voltage-reactive power curve reflecting the constraint according to the following equation.

(N _const is the constraint setting, w _c is the penalty coefficient, PF _c is the weight for each constraint)

Images