JP6682402B2 - Power system power flow value estimation method and apparatus - Google Patents

Description

  The present invention relates to a method and apparatus for estimating a value related to power flow in a power system.

  Due to the rapid acceleration of global warming in recent years, social demands for increasing the amount of solar power generation are increasing. With the increase in the amount of solar power generation, the actual load can be known when the amount of solar power generation can be grasped even in units such as distribution banks in distribution substations, which is advantageous for system control. In addition, it is expected that it will lead to understanding the actual situation of the advanced power factor problem of distribution systems that has become apparent in recent years.

  The power factor is an important input for estimating the amount of solar power generation, but there are many places where the power factor is not measured for each distribution bank of a distribution substation. At the distribution substation, the electric power obtained from the upper substation is branched and sent to multiple distribution banks.At this time, the measured values of the electric power, current, and voltage from the upper substation are grasped as vector values. However, it is often the case that the measured values of electric power, current, and voltage are scalar values, for example, only the effective values are known for each distribution bank of the distribution substation. Therefore, there is a demand for a method of estimating the power factor from the effective values of current and voltage in the distribution system.

  In this regard, Patent Document 1 proposes a method of estimating a power flow value such as active power or reactive power of a power system from a limited number of measurement values. The method of Patent Document 1 obtains a relational expression of active power and reactive power with respect to apparent power in advance and calculates active power and reactive power from the measured value of apparent power using the relational expression at the time of estimation.

JP, 2007-166870, A

  In the method of Patent Document 1, the condition is that the values of active power and reactive power with respect to apparent power can be sufficiently described by a relational expression such as a linear function. However, this condition does not hold when a distributed power source such as photovoltaic power generation is connected to the system whose power flow is to be estimated. This is because the output of a distributed power source such as photovoltaic power generation fluctuates greatly in a short time due to weather conditions and the like, and the values of active power and reactive power with respect to apparent power cannot be uniquely obtained from the relational expression.

  Therefore, it is an object of the present invention to provide a method of estimating a power flow value from a limited type of measurement information even in a system in which a distributed power source such as a photovoltaic power generation system is introduced. In particular, in the configuration of the system where the tidal current value is measured on the upstream side of the branch of the system, but only the scalar values (effective value, etc.) of voltage and current are measured for each branch line on the downstream side of the branch, A method for estimating each power flow value (vector value of current) after branching is provided.

  In order to solve the above problems, a typical power flow estimation method of the present invention is to input and store a current in consideration of a phase with respect to a voltage as a vector value from the upper side of a branch of a power system, and store each branch on the lower side of the branch. Input and store the current as a scalar value from the line, select one of the multiple branch lines on the lower side of the branch as the target for power flow estimation, and check the current scalar value of the branch line other than the selected branch line. A past time having a past value that is substantially equal to the value is obtained, and for the vector value of the current on the upper side of the branch, a difference vector of the vector values at the current time and the past time is obtained. From the current scalar value at the past time and the current time, find the candidates for the deviation angle of the current value of the branch line for the power flow estimation, narrow down the candidates for the deviation angle using the statistical property of the deviation angle of the current value, and estimate the power flow. From the value of the narrowed declination of current elephant branch line, characterized in that to obtain the physical quantity corresponding to the power flow value.

  Further, one example of the present invention measures the current as a vector value in consideration of the phase with respect to the voltage on the upper side of the branch of the power system, and the power flow value of the power system measuring the current as a scalar value on the lower side of the branch. An estimation method, the first step of obtaining a current vector value on an upper side of a branch of a power system and a current scalar value on a lower side branch line of the branch, and one of a plurality of branch lines on a lower side of the branch. Is selected as a target of power flow estimation, a third step of obtaining past time cross-section data that is substantially equal to the current value for the scalar value of the current in branch lines other than the selected branch line, and the current step The fourth step of calculating the difference vector between the vector value of the current on the upper side of the branch and the vector value of the current on the upper side of the current branch in the past time section which is substantially equal to the value, and the difference vector and the difference vector. And a fifth step of obtaining a candidate for a deviation angle of the current value of the branch line of the power flow estimation using a triangle composed of a past time section of the branch line of the power flow estimation and a scalar value of the current The sixth step of narrowing down the candidates for the angle of deviation using the statistical property of the angle of deviation, and the step of converting the value of the angle of deviation of the current of the branch line on the lower side of the narrowed branch into a required physical quantity such as a power factor value. It is characterized by comprising 7 steps.

Further, according to the present invention, a measured value storage in which a current considering a phase with respect to a voltage is input and stored as a vector value from the upper side of a branch of a power system and a current is input and stored as a scalar value from each branch line on the lower side of the branch is stored. Section, a declination estimation target bank selecting section that selects one of a plurality of branch lines on the lower side of the branch as a target of power flow estimation, and a current value of a scalar value of current in branch lines other than the selected branch line. A past time point determination unit that obtains a past time point having a past value that is approximately equal to
For the vector value of the current on the upper side of the branch, a difference vector calculation unit that obtains a difference vector between the vector values at the current time and the past time, and the difference vector, and a scalar of the current at the past time and the current time of the branch line for which the power flow is to be estimated. From the value, a candidate creation unit that obtains candidates for the deviation angle of the current value of the branch line for which the power flow is to be estimated, a declination candidate narrowing unit that narrows down the declination candidates using the statistical properties of the declination of the current value, and the tidal current A power flow value estimation device for a power system, comprising a conversion unit that obtains a physical quantity corresponding to a power flow value from a value of a narrowed declination of a current of a branch line to be estimated.

  According to the present invention, even when the measurement data at the branch line to be estimated is the scalar value of the current, the deviation angle of the current can be estimated, and the power flow value such as active power or reactive power can be known. Problems, configurations, and effects other than those described above will be clarified by the following description of the implementation mobile phone.

The figure which shows an example of the typical power transmission and distribution system structure to which this invention is applicable. The simplified figure for demonstrating the method of power flow value estimation. The figure which shows the vector relationship of Formula (1). Diagram for explaining the possible values of the declination candidate vector _{I 1, I 2, I 3} . Shows an example of a candidate of the argument of the vector _{I 1, I 2, I 3} . The figure which shows the analysis result of the property of the scalar which can be used as a measured value of the lower side of a branch. The relationship between the vectors _{I 0} and the vector _{I 1, I 2, I 3} , illustrates provisionally set. The figure which shows the relationship of two vectors in two different time. Figure it is assumed that current and past values of the vector I ₂ and vector I ₃ are equal. Diagram showing the relationship between vector I ₀ and the vector I ₁ when the Figure 7c is established. The figure which shows the state which is comparing the vector relationship of the past time with respect to the vector relationship of the present time (t = tn). The figure which plotted the time width t _{diff (tn)} which goes back in the past from tn of 10 am to 14:00. The figure explaining the determination method 1 of a declination candidate. The figure explaining the determination method 1 of a declination candidate. The figure explaining the determination method 1 of a declination candidate. The figure explaining the determination method 1 of a declination candidate. The figure explaining the determination method 2 of a declination candidate. FIG. 3 is a block diagram of a power flow value estimation device. The flowchart of the power flow value estimation method.

  Embodiments of the present invention will be described in detail below with reference to the drawings.

  FIG. 1 is a diagram showing an example of a typical power transmission and distribution system configuration to which the present invention is applicable. In the power transmission / distribution system, the power transmission system is a part from the upper substation 102 on the upper side 111 to the distribution substation 103 to which the present invention is applied via the power transmission line 104. In these upper system parts, various quantities such as active power P, reactive power Q, voltage V, and current I are set as the output side measured value 120 of the upper substation 102 or the input side measured value 121 of the distribution substation 103. It is measured as a vector quantity. Therefore, the power factor can be measured or estimated at these points on the power transmission system.

  The distribution substation 103 is composed of a plurality of banks via a transformer 108. A large-scale customer load 106, a small-scale customer load 107, and a distributed power source 110 such as solar power generation or wind power generation are connected to the power distribution system and reach the lower side 112. Regarding such a distribution substation 103, measurement of various quantities such as active power P, reactive power Q, voltage V, and current I in a branched lower distribution line is often performed with voltage V and current I as scalar quantities. There are many places where the power flow values such as the active power P and the reactive power Q are not measured. Therefore, the power factor cannot be obtained at these points in the distribution system after branching. The distribution line may be branched without a transformer.

  In this way, generally, on the upper side, since the number of points to be measured is small, it is possible to measure power flow data such as active power and reactive power as a vector quantity. However, the number of measurement points increases as the branching is repeated toward the lower side. Therefore, in many cases, the voltage or current is measured with a scalar value (effective value) because of cost or the like.

  On the other hand, distributed power sources such as photovoltaic power generation, which have been introduced in recent years, can often be located near the end of the distribution system. In order to control the power transmission and distribution system such as grasping the actual load at the time of reclosing, there is a demand for accurately grasping the power generation amount of the distributed power sources. However, it is difficult to grasp the tidal current because the tidal current is not measured near the end of the distribution system to which solar power generation is connected and the current measurement is often a scalar value. In addition, the relaxation of restrictions on reverse power flow on a bank-by-bank basis has further increased the need to understand the power flow at the end of the distribution system.

  Therefore, there is a demand for development of a method capable of estimating a tidal current in a bank unit by diverting an existing current sensor that measures a scalar value as much as possible. Therefore, in the present invention, in the distribution substation of FIG. 1, the active power as the power flow value 123 of the estimation target on the lower side of the branch is determined from the scalar quantity of the voltage V and the current I which are the measured values 122 on the lower side of the branch. The vector values of P and reactive power Q are estimated. This means obtaining the power factor for the distribution system after branching.

  In the following description, the estimated target power flow value 123 on the lower side of the branch to be obtained is simply referred to as a power flow value. However, this is at least one of the active power value and the reactive power value, or the power factor or the current with respect to the voltage. It is a concept that includes a declination, a vector value of current, and the like. Since these can be converted into each other, obtaining them can be regarded as essentially the same as obtaining the power flow value 123 of the estimation target on the lower side of the branch. Further, the present invention can be applied to the apparent electric power value on the branch side almost similarly as a scalar value of the current on the downstream side of the branch. In addition, although the transformer 108 is shown on the lower side of the branch in FIG. 1, the present invention can be similarly applied to a branch without the transformer 108. Further, the present invention can be applied not only to a distribution substation but also to a branch of another layer.

  In the following description of the present invention, the case where the power flow value is the vector value of the current will be described as an example.

FIG. 2 shows a simplified diagram for explaining the method of power flow value estimation. Here, the values measured on the input side of the distribution substation 103 and on the plurality of branched output sides (banks 1, 2, 3) are displayed. The incoming-side measured value 121 of the distribution substation 103 is a vector value having an absolute value of the current and a deviation angle with respect to the voltage, which is expressed as I ₀ . The measured value 122-1 of the bank 1 is a scalar value having an absolute value of current, which is | I ₁ |. Similarly, the measured values 122-2 and 122-3 of the banks 2 and 3 are also scalar values having the absolute value of the current, which are expressed as | I ₂ | and | I ₃ |. In the present invention, the bank 1, the measured value in the form of the scalar value | I 1 _| from obtaining the estimate I ₁ as a vector value 141. The same applies to banks 2 and 3.

  FIG. 2 focuses on branch points in the distribution substation 103. In the same figure, in order to facilitate subsequent calculations, the same voltage value is converted and shown. For example, when the current measurement for each bank is performed on the secondary side of the transformer 108, it is converted to the current value on the primary side facing the branch. For example, when the transformer 108 is converting 66 [kV] /6.6 [kV], the current value converted to the primary side is 1/10 of the measured value on the secondary side. Further, a more detailed conversion may be performed by reflecting the characteristics in consideration of the equivalent circuit of the transformer.

  Also, although the power flow value is measured on the upper side of the branch, it is sufficient to describe it as a vector value of the current because it has the same voltage (there is an absolute value of the current and a deviation angle with respect to the voltage). When the power flow measurement point on the upper side of the branch is the sending point of the upper side substation, it can be converted to the value 121 immediately before the branch by multiplying the impedance of the transmission line 104 of the system. The measured value of the current on the lower side of the branch is a scalar value, but when the above conversion is performed, according to the present invention, it is possible to reduce the problem of estimating the declination of the current on the lower side.

By the above conversion, the power factor estimation in the method of the present invention makes the vector I ₀ and the scalars | I ₁ |, | I ₂ |, | I ₃ | known, and declinates the vectors I ₁ , I ₂ , I ₃ . Is equivalent to estimating If the deflection angle of the current is obtained, it can be easily converted into a power factor, active power P, reactive power Q value, or the like. Of course, it may be described in terms of apparent power and argument, active power and reactive power, and the like. In addition, although an example of three branches is shown in the present embodiment, generally n (n: a natural number of 2 or more) branches can be similarly applied.

Next, the problems faced when estimating the argument from the scalar value of the current will be described. First, regarding the branch portion of the distribution substation 103 in FIG. 2, the expression (1) is established.
[Equation 1]
I ₀ = I ₁ + I ₂ + I ₃ (1)
FIG. 3 is a diagram showing the vector relationship of the equation (1). This is because when scalars | I ₁ |, | I ₂ |, and | I ₃ | that are the lengths of the vectors are given, an example of a combination of the vectors I ₁ , I ₂ , and I ₃ satisfying the equation (1) is This indicates that the result is as shown in FIG. In FIG. 3, it is drawn horizontally with the declination of I ₀ as a reference. If the declination of I ₁ , I ₂ , and I ₃ can be correctly estimated, the sum of the vector I ₁ , the vector I ₂ , and the vector I ₃ becomes equal to the vector I ₀ . Actually, the magnitudes | I ₁ |, | I ₂ |, and | I ₃ | of I ₁ , I ₂ , and I ₃ are determined from the measured values.

FIG. 4 is a diagram for explaining possible values of the argument candidates of the vectors I ₁ , I ₂ , and I ₃ . Here, as shown in FIG. 4, when circles C ₁ , C ₂ , and C ₃ having radii I ₁ , I ₂ , and I ₃ , respectively, are considered, candidates for the declination angles of I ₁ , I ₂ , and I ₃ are as follows. Like First, the center of the circle C ₁ is placed at the start point of the vector I ₀ , then the center of the circle C ₂ is placed on the circumference of the circle C ₁ , and the center of the circle C ₃ is placed at the end point of the I _0. There are candidates for the declination of I ₁ , I ₂ , and I ₃ when C ₂ and the circle C ₃ have an intersection. Specifically, the vector _{I 1} is (two generally) from the center of the circle _{C 1} the center of the circle _{C 2,} vector _{I 2} is the intersection of circle _{C 2} and the circle _{C 3} from the center of the circle _{C 2,} vector _{I 3} is From the intersection (generally two) of the circle C ₂ and the circle C ₃ , the vector becomes the start point and the end point at the center of the circle C ₃ , respectively. Incidentally, place the circle on the center of the circle C ₂ of the circle C _3, the same also obtain the intersection of circle C ₂ and the circle C _1. Further, the same applies even if the order of the circle C ₁ , the circle C ₂ , and the circle C ₃ is arbitrarily changed.

FIG. 5 is a diagram showing an example of candidates for argument angles of the vectors I ₁ , I ₂ , and I ₃ . As can be understood from FIG. 4, since these combinations are innumerable as shown in FIG. 5, it is understood that the equation (1) alone is not sufficient to estimate the power factor.

  Therefore, it is necessary to use some additional information or make assumptions based on the characteristics peculiar to the power system. As an example of the former, it is conceivable to use external information such as the addition of measurement values by adding sensors and the amount of solar radiation. However, it is disadvantageous in that new capital investment is required and that operating costs increase. On the other hand, the latter may be realized only by the algorithm. Therefore, in the present embodiment, the latter method that uses only the assumption based on the characteristics peculiar to the power system is used.

As a characteristic peculiar to the power system, the relationship between the input and the output in power factor estimation is used. The input here is input data used in power factor estimation. In the present invention, it refers to the measured value (scalar | I ₁ |, | I ₂ |, | I ₃ |) of the current on the lower side of the branch. The output is an estimated value of the power factor, and specifically refers to the declination of the vectors I ₁ , I ₂ , and I ₃ .

FIG. 6 shows the result of analysis of the properties of the scalars | I ₁ |, | I ₂ |, | I ₃ | (size, effective value, etc.) that can be used as the measurement values on the lower side of the branch. Here, the relationship between the scalars | I ₁ |, | I ₂ |, | I ₃ | and their deflection angles is shown. This figure is a scatter diagram in which the horizontal axes represent the scalars | I ₁ |, | I ₂ |, and | I ₃ |, and the vertical axes represent the deviation angles of the corresponding vectors I ₁ , I ₂ , and I ₃ . The plotted time period is one hour from 10 am to 11 am for a day.

From the analysis result of FIG. 6, it can be seen that there is a certain degree of correlation between the scalars | I ₁ |, | I ₂ |, | I ₃ | and their declinations. FIG. 6 shows measured values of a system in which photovoltaic power generation is connected as a distributed power source to each of the three lower branches. Since the figure is data of the width of the time when the load on the customer can be regarded as almost constant, it is estimated that most of the fluctuations in the current value are caused by solar power generation. Fluctuations due to solar power generation vary according to the power factor of the converter in the power generation facility. Therefore, as shown in FIG. 6, the plot of the deviation angle with respect to the magnitudes of the scalars | I ₁ |, | I ₂ |, | I ₃ |

From the above relationship, the magnitudes of the scalars | I ₁ |, | I ₂ |, | I ₃ | and the declination angles of the vectors I ₁ , I ₂ , I ₃ can be calculated in a short time such as one hour shown in FIG. If so, it can be seen that there is a relationship that can be regarded as one price. It should be noted that in the present invention, “single price” means that when a certain value on the horizontal axis is determined, one value on the vertical axis is correspondingly determined.

If the relationship between the magnitudes of these scalars | I ₁ |, | I ₂ |, | I ₃ | and their declinations is always constant, an approximate expression is created to obtain the declination angles of I ₁ , I ₂ , and I ₃ . Will be required and the problem will be solved. However, actually, the load of the customer is superimposed on the measurement values at the measurement points of I ₁ , I ₂ , and I ₃ . Therefore, a different trajectory is exhibited for each time zone, each day type such as weekdays and holidays, or each season. In addition, new / discontinued solar power generation sites and suspension of inspections due to inspections can also be factors that cause different paths.

Therefore, in the present invention, it is decided to use the property with less requirements in relation to the magnitudes of the scalars | I ₁ |, | I ₂ |, | I ₃ | and their deviation angles. Specifically, it is established when each deviation angle can be regarded as a single value with respect to the sizes of the scalars | I ₁ |, | I ₂ |, and | I ₃ |. This relaxed requirement eliminates the need to consider factors such as distinction by time of day, weekdays on holidays, and seasons.

Further, the reason why each of the deviation angles with respect to the sizes of the scalars | I ₁ |, | I ₂ |, | I ₃ | As described above, the photovoltaic power generation, which is the main factor of the temporal fluctuation of I ₁ , I ₂ , and I ₃ , is interconnected with a predetermined power factor. Therefore, the plots of the scalars | I ₁ |, | I ₂ |, and the respective deviation angles with respect to the magnitude of | I ₃ | exhibit a substantially constant locus. In addition, the load fluctuation of the customer can be considered to be small within a certain time such as 1 hour or 30 minutes, and therefore the deviation with respect to the size of the scalars | I ₁ |, | I ₂ |, | I ₃ | The effect can be neglected as to whether or not the relationship with the horn can be regarded as almost monovalent. The property that the load can be regarded as constant within a fixed time such as 1 hour or 30 minutes is used as the setting of the maximum value t _diffMax in the description of FIG. 8 described later.

Next, an implementation method according to an embodiment of the present invention will be described with reference to FIGS. 7a, 7b, 7c, and 7d. In the present embodiment, as described above, the fact that the relationship between the magnitudes of the scalars | I ₁ |, | I ₂ |, | I ₃ | The declination of the vectors I ₁ , I ₂ , and I ₃ is estimated from the time change of I ₀ .

First, FIG. 7A is a diagram temporarily showing the relationship between the vector I ₀ and the vectors I ₁ , I ₂ , and I ₃ . At this stage, the declination angles of the vectors I ₁ , I ₂ , and I ₃ are all indefinite, and it is only known that the vector sum of these deviates from the vector I ₀ . The vector relationship of FIG. 7a satisfies the relationship of Expression (1) described above.

Next, FIG. 7b illustrates two vector relationships at two different times. One vector relationship is the situation of Figure 7a, which is the vector relationship at the current time tn. Another vector relationship is a vector relationship at a certain time t ₀ in the past. As shown in FIG. 7b, it is assumed that the vector relationship between the current time tn (t now) and a certain time point to (t old) in the past is as illustrated. Incidentally, the vector _I 0 at time _{tn, I 1, I 2,} I 3 and _{I 0tn, I 1tn, I 2tn} , I 3tn and vector _I 0 at time _{to, I 1, I 2,} I 3 and _{I 0to, I 1to, I 2to,} it has been described as _{I 3to.}

At each of the time tn and the time to, the vector sums shown in the equations (2) and (3) hold. Here, it is assumed that the certain time to in the past is not necessarily one unit time before the current time tn. This equation, since at each time section (1) is _satisfied, I _1tn, I _2tn, the end point of the synthesis vector and _{I 0Tn} of _{I 3Tn} is consistent ((see 2)). Similarly _{I 1to, I 2to,} the end point of the synthesis vector and _{I 0To} of _{I 3To} is consistent ((3) refer to formula).
[Equation 2]
I _0tn = I _1tn + I _2tn + I _3tn (2)
[Equation 3]
I _0to = I _1to + I _2to + I _3to (3)
FIG. 7c is a diagram assuming that the current value and the past value of the vector I ₂ and the vector I ₃ are equal. Then if, as in (1) in FIG. 7c, substantially equal vector _{I 2TN} and vector _{I 2To,} and as shown in (2) of FIG. 7c, when the substantially equal condition holds true vector _{I 3Tn} and vector _{I 3To} I assume. That is, for _{I 2,} there is substantially equal past values _{I 2To} and the measured current value _{I 2TN,} and for _{I 3,} and that approximately equal past values _{I 3To} and the measured current value _{I 3Tn} exist To do.

FIG. 7d is a diagram showing the relationship between the vector I ₀ and the vector I ₁ when the condition of FIG. 7c holds. In this case, [Delta] I a difference vector of the vector _{I 0Tn} and vector _{I 0to 0tn,} and when the difference vector of the vector _{I 1Tn} and vector _{I 1TO} defined as ΔI _1tn, as (3) in FIG. 7d, a difference vector [Delta] I _0Tn The difference vector ΔI _1tn is almost equal.

Here, since the deviation angles of the vector I _0tn and the vector I _0to are known, the deviation angle of the difference vector ΔI _0tn is also known. Therefore, the deviation angle of the difference vector ΔI _1tn is known, and it is possible to obtain the deviation angle of the vector I _1tn from the triangle formed by the scalar | I _1tn | and the scalar | I _1to | and the difference vector ΔI _1tn . Become. There are generally two possible candidates for the argument of the vector I _1tn , and the selection method for these will be described later.

An assumption is made in the method of calculating the argument of the vector I _1tn described above. Specifically, the _assumption is that the vector I _2tn is approximately equal to the vector I _2to and the vector I _3tn is approximately equal to the vector I _3to . That is, assuming that the _{I 2,} there is substantially equal past values _{I 2To} and the measured current value _{I 2TN,} and for _{I 3,} approximately equal past values _{I 3To} and the measured current value _{I 3Tn} exist Is.

The conditions under which this assumption holds will be described. At each stage of FIGS. 7a, 7b, and 7c, the declination angles of the vectors I ₁ , I ₂ , and I ₃ are all indefinite. At the stage of FIG. 7d, the declination of the vector I ₁ can finally be determined (if the above assumption is correct). Therefore, in all the states shown in FIGS. 7a, 7b, 7c, and 7d, the declination angles of the vectors I ₂ , I ₃ (including I _2tn , I _2to , I _3tn , I _3to ) remain indefinite. .

Now, let us consider the requirement that the relationship between the magnitudes of the scalars | I ₁ |, | I ₂ |, and | I ₃ | and the declination described above with reference to FIG. The requirement on the left shows that when the sizes of the scalars | I ₂ | and | I ₃ | are determined, the respective approximate declinations can be obtained. Therefore, if the relation that the scalar | I _2tn | and the scalar | I _2to | are substantially equal to each other, the respective declination angles are approximately equal to each other, and the vector I _2tn and the vector I _2to are approximately equal to each other. . The same applies to I ₃ . Therefore, even when the declinations of the vectors I ₂ and I ₃ are treated as indefinite from beginning to end, when the scalar | I _2tn | and the scalar | I _2to | are almost equal and the scalars | I _3tn | and the scalar | I _3to | , And the vector I _2tn and the vector I _2to are substantially equal to each other, and the vector I _3tn and the vector I _3to are substantially equal to each other, as shown in (1) and (2) of FIG. 7c.

From the above, the condition that (vector I _2tn + vector I _3tn ) and (vector I _2to + vector I _3to ) are substantially equal holds, and both are substantially the same vector at both time tn and time to (time to and time tn. It does not change in time). Therefore, it can be considered that the condition that the difference vector ΔI _0tn and the difference vector ΔI _1tn are substantially equal to each other is satisfied as shown in (3) of FIG. _7D . It should be noted that if the roles of I ₁ , I ₂ , and I ₃ are sequentially switched, the declination of the vectors I ₂ and I ₃ can be obtained.

Next, _referring to FIG. 8, consider the condition that the scalar | I _2tn | and the scalar | I _2to | are substantially equal to each other, and the scalar | I _3tn | and the scalar | I _3to | are approximately equal to each other. To determine accurately the argument of the vector _{I 1} in the above-described manner, the scalar _{| I 2TN} | scalar _{| I 2To} | of the difference, and the scalar _{| I 3tn |} scalar _{| I 3to |} as the difference is small good. However, the difference does not always become small at time tn−1 immediately before time tn. This is because the amount of change in the amount of solar power generation over time is large. This is because in solar power generation, there are some cases in which the output fluctuates by more than half of the rated value in a short time due to an increase or decrease in the amount of solar radiation, and there is a possibility that the output fluctuates sufficiently during the sampling interval.

FIG. 8 is a diagram showing a state in which the vector relationship at the past time is compared with the vector relationship at the current time (t = tn). Therefore, as shown in FIG. 8, the conditions that the scalar | I _2tn | and the scalar | I _2to | are substantially equal to each other and the scalar | I _3tn | and the scalar | I _3to | are approximately equal to each other are traced back from time tn. Calculate time to.

Note that FIG. 8 shows the state in which the horizontal axis represents time and the vector relationship of the past time is compared with the vector relationship Vn of the current time (t = tn) shown in FIG. 7b. For example, with respect to the vector relationships Vn-1 and Vo between time tn-1 and time to, it means to search for a vector in which the magnitudes of I ₂ and I ₃ are substantially the same at the same time.

As the selection criterion I _2diff here, the value of the equation (4) is 0.1 or less. More preferably, it may be 0.05 or less.
[Equation 4]
I _2diff ≡ | {(scalar | I _2tn |)-(scalar | I _2to |)} / [{(scalar | I _2tn |) + (scalar | I _2to |)} / 2}] | ≦ 0.1 ( 4)
The reference _{I 3Diff} selection, (5), as shown in the expression is the same as the reference _{I 2Diff} selection.
[Equation 5]
I _3diff ≡ | {(scalar | I _3tn |)-(scalar | I _3to |)} / [{(scalar | I _3tn |) + (scalar | I _3to |)} / 2}] | ≦ 0.1 ( 5)
The past time to when these conditions are simultaneously satisfied for I ₂ and I ₃ and I _2diff + I _3diff is minimum is obtained. At this time, since the time width t _diff ≡tn-to that goes back to the past changes depending on the current time tn, it is expressed as t _{diff (tn)} as a function of _tn .

At this time, for example, 1 hour or 30 minutes is selected as the maximum value t _diffMax of the time width that goes back in the past. t _diffMax is a time width over which the load on the consumer can be regarded as substantially constant. This is a time width in which the respective deviation angles with respect to the magnitudes of the scalars | I ₁ |, | I ₂ |, and | I ₃ | in FIG. If it is attempted to determine the retrospective time to the time exceeding the time width shown on the left, the load on the customer cannot be considered to be constant, and the declination angle for the magnitudes of the scalars | I ₁ |, | I ₂ |, | I ₃ | It cannot be regarded as a single price.

Since t _diffMax is an amount related to the fluctuation amount of the load of the customer, t _diffMax is large from around 9:00 am to noon, or around 13:00 to around 16:00, when the load of the customer can be regarded as almost constant. Good as a value. Similarly at night, t _diffMax can be increased.

On the contrary, t _diffMax may be reduced from around 6:00 am when the load of the consumer rises to around 9:00 am. Similarly, t _diffMax may be reduced in a time zone in which the fluctuation of the customer load is large in the evening or after noon.

FIG. 9 is a plot of the time width t _{diff (tn)} traced back in the past, where tn is from 10 am to 14:00 am. Here, t _diffMax is fixedly 1 hour (3600 seconds). It can be seen from the figure that t _{diff (tn)} has a rough tendency to increase in a sawtooth shape that rises to the right. This is considered to reflect that the condition that the scalars | I ₂ | and | I ₃ | have the same magnitude at a certain point in the past occurs intermittently. Further, when tn is about 30 minutes from noon, t _{diff (tn)} has a small value. It is expected that this is because the load on the customer due to lunch break will decrease, and the conditions for the latest scalars | I ₂ | and | I ₃ |

On the other hand, after 12:30 in the daytime, the load on the customer starts to increase again, and by following the path opposite to the gradual decrease in the load on the customer after noon, there is a tendency that t _{diff (tn)} increases. However, the serrated inclination at this time is steeper than in normal times. This is because the point where the load gradually decreases after noon is traced in reverse, and when tn increases by 1, t _{diff (tn) tends to} increase by 1 or more (becomes larger than the slope of 1 in normal times).

After 13:00, the state of the consumer load during the lunch break ends, so that the saw-toothed t | I ₂ | and | I ₃ | trends with a slope of 1 start again.

Next, the declination candidate determination method 1 will be described with reference to FIGS. 10a, 10b, 10c, and 10d. In FIG. 7d described above, it has been explained that there are generally two candidates for the argument when _obtaining the argument for the vector I _1tn . This is the same as when _obtaining the I _1to argument, and there are two argument candidates. Further, the same applies to I ₂ and I ₃ , and there are generally two candidates for the deviation angles of I _2tn and I _3tn .

In the following, an idea for narrowing down this will be described. This procedure utilizes the property that the vector sum of the vectors I ₁ , I ₂ , and I ₃ is equal to the vector I 0. For each of the vectors I ₁ , I ₂ and I _{3, there} were declination candidates [I ₁₊ , I ₁₋ ], [I ₂₊ , I ₂₋ ], [I ₃₊ , I ₃₋ ] ([. ] Represents an alternative)). That is, the vector _{I 1 I 1+} or _{I 1-exist, I 2+} or _{I 2-exist} on the vectors _{I 2, I 3+} or _{I 3-} is present on the vectors _{I 3,} the vector _{I 4 I 4+} or I ₄₋ is present.

In this case, there are eight end points by the combination of vector sums, A, B, C, D, E, F, G, and H shown in FIGS. 10a, 10b, 10c, and 10d. FIG. 10a shows a case where I ₁₊ and I ₂₊ are changed to I ₃₊ or I _3−, and finally points A and B are indicated. FIG. 10b shows an example in which I ₁₊ and I _{2− are} sometimes I ₃₊ or I _3−, and finally points C and D are indicated. FIG. 10c shows a case where I ₁₋ and I ₂₊ are changed to I ₃₊ or I _3−, and finally points E and F are indicated. FIG. 10d shows an example of I ₁₋ , I _2−, and sometimes I ₃₊ or I _3−, and finally points G and H.

Those in which these end points A, B, C, D, E, F, G, and H are included in the vicinity of the end point of I ₀ (described as the reference region in FIG. 10a) are the combinations of correct declination candidates. However, in this method, as shown in FIGS. 10a and 10b, when both of the end points B and C fall within the reference area near the end point of I ₀ , it is not possible to narrow down the candidates for the declination angle. In this way, when the end points B and C are candidates, it is impossible to narrow down the two vectors [I ₂₊ , I ₂₋ ] and [I ₃₊ , I ₃₋ ]. Further, when the variation amount of I ₁ , I ₂ , and I ₃ is large and it is necessary to increase the reference area, the end point E also becomes a candidate as shown in FIG. 10c, and [I ₁₊ , I _1- ] You can no longer narrow down. Therefore, it is assumed that the method of comparing the end points of the vectors shown in FIGS. 10a, 10b, 10c, and 10d may be difficult to narrow down depending on conditions.

Therefore, an example of the declination candidate determination method 2 will be described with reference to FIG. On the right side of FIG. 11, the I ₁₊ and I _1-candidates polarization angle for I _1, is shown in time series by the dotted and solid lines. Similarly, the thick dotted line on the right side of the figure is the correct value of the argument of I ₁ . The declination of I ₁₊ and I ₁₋ varies frequently, as shown. However, in order to switch the mutual roles, the case where the argument is close to the correct value and the case where the argument is apart from the correct value are taken. On the other hand, the time series change of the declination angle of I ₁ , which is the correct value, is often relatively smaller than the declination angles of I ₁₊ and I ₁₋ . Therefore, the histogram of the declination angles of I ₁₊ and I ₁₋ for a short time is calculated while sliding the time axis. When the two histograms are added together, the frequency of the class close to the correct value of the argument of I ₁ increases. Therefore, a plurality of argument candidates can be narrowed down by selecting the argument angles of I ₁₊ and I ₁₋ depending on whether or not they are close to the mode of the histogram on the left. In this method, compared with the time variation of the declination candidate, the time variation of the correct value of the declination is relatively small, and the value when the declination candidate is incorrect varies relatively, so that a specific peak Utilized the properties that are difficult to make.

The feature of this narrowing-down method is that narrowing-down can be performed only from the declination candidates of one vector. For example, when narrowing down the candidates for the argument of I ₁ , the information about the candidates for the arguments of I ₂ and I ₃ is unnecessary. Due to this property, it is possible to prevent the narrowing down of two or more declination candidates in a chained manner as in the schemes shown in FIGS. 10a, 10b, 10c, and 10d.

  FIG. 12 shows a block diagram of the power flow value estimation device. Here, the lower branch line after branching is described as a bank, but the same can be applied to branches of other layers in which the relationships of the types of measured values are similar.

In the power flow value estimation device of FIG. 12, first, the measured value 312 is acquired by the measured value acquisition unit 313 and stored in the measured value storage unit 321. The measured value 312 here includes at least the vector I ₀ and the scalars | I ₁ |, | I ₂ |, | I ₃ |, and also includes the voltage values on the upper side and the downstream side.

  Next, the measurement value conversion unit 314 simplifies the process corresponding to FIG. This is to simplify the problem of determining the deviation angle of the current vector by calculating the same voltage value before and after the branch. The operation of the measurement value acquisition unit 313 and the measurement value storage unit 321 or the operation including the measurement value conversion unit 314 may be performed asynchronously with the subsequent power flow estimation operation. For example, the measurement data may be stored or converted as soon as they arrive.

  Next, the operation of the blocks related to the series of operations of the power flow estimation will be described. These series of estimation operations are sequentially processed according to an instruction from the overall control unit 318. First, the estimation target bank selection unit 315 selects a bank whose tidal current is estimated. For example, of the banks 1, 2, and 3 in FIG. 2, first, the tidal current (such as the power factor) of the bank 1 is set as an estimation target.

  Next, the past time point to decision unit 316 decides the past time point to to be subjected to the time difference processing. The method of obtaining the time to is as described in the description of FIG. 8 above. When determining the past time section to, the threshold value for considering the scalar values of the currents to be compared as substantially the same value can be appropriately changed by the set value input unit 317 such as the threshold value. When the fluctuation of the current of the system to be estimated is large, the set value input unit 317 performs an operation such as increasing the threshold value for considering the current values to be the same. Next, the time difference creation unit 322 creates a time difference. Specifically, this block calculates the vector difference (vector value) of the current vector on the upper side of the branch.

  Next, the declination candidate creation unit 323 creates declination candidates (generally two) at time tn. Specifically, a triangle is constructed from the vector value of the difference created by the block of the time difference creating unit 322 and the scalar value of the current of the estimation target bank at the current time tn and the past time to, and the deviation of the current time tn is calculated. Find a corner candidate. Further, since the declination candidates of the past time section to are also obtained, the declination estimation processing of the time sections tn and to may be performed at the same time.

  Next, the declination candidate narrowing unit 324 selects an appropriate candidate from the generally two declination candidates created by the declination candidate creation unit 323. As the selection method, the method described in FIG. 11 can be used. In the case of the estimation method that uses only the portions where the argument values of the plurality of argument candidates are substantially the same, the argument candidate narrowing unit 324 of this block is replaced with the identity determination unit of the argument value of the argument candidates. In the case of narrowing down the candidates using the method of FIGS. 10a, 10b, 10c, and 10d, the candidates of the argument are similarly calculated for all the other banks, and then the selection is performed.

  Next, the conversion unit 325 from the declination angle to the physical quantity of the usage pattern converts the physical quantity to the target physical quantity. For example, it is converted into a power factor value, active power value, reactive power value, or the like. The result of conversion or the deflection angle value of the current before conversion is output from the estimation result output 329.

  After the series of estimation processes is completed, if the estimation target bank selection unit 315 sequentially sets the estimation target of the power flow to the next bank, the power flow values of all the banks can be estimated.

It should be noted that the following modifications may be further made in executing the processing of FIG. The measurement value storage unit 321 may store the measurement value as it is. Alternatively, the vector value of the current converted by the measurement value conversion 314 (upper side of the branch) and the scalar value of the current (lower side of the branch) may be stored. Further, the storage period is sufficient if it is at least t _diffMax .

  In the measurement value conversion 314, in order to simplify the problem of determining the deviation angle of the current vector in simplification corresponding to FIG. 2, if the conversion of the voltage or the like is performed each time, this conversion is not necessarily performed. good. For example, the present invention can be similarly applied even if the scalar value of the current on the lower side of the branch is calculated with the voltage and calculated as the apparent power value. In this case, the upper side of the branch may be one that calculates not by the vector value of the current but by the apparent power and its deviation angle (or active power and reactive power).

  Next, the flow of the power flow value estimation method will be shown using FIG. Processing steps S414 to S417 are a flow of the measurement value storage task. These may be performed asynchronously or synchronously with the task of power flow estimation. First, in process step S415, a measurement value is acquired. In this case, the sensor information (not shown) may be directly input through communication or through input means (not shown) such as A / D conversion. In process step S416, the measured value is converted. This is done for simplification as a problem of current declination estimation corresponding to FIG. 2, but it may be converted before being stored in the measurement value storage unit as in this flowchart, or processing steps S436 and S437 described later. When reading from the measurement value storage unit 321, the conversion may be performed at any time. The data is stored in the measurement value storage unit 321 after conversion as necessary (processing step S417).

  Next, the flow of the tidal current estimation task will be described using the processing steps S432 to S440. First, in process step S432, execution of the tidal current estimation task is started. In processing step S433, the current time section tn is sequentially updated. In the processing step S434, a loop processing is performed in which the processing procedure from the processing steps S435 to S439 is repeatedly executed while changing the target bank.

  In the iterative process, in process step S435, one of the banks on the lower side of the next plurality of branches is selected as the target of power flow estimation. Next, in processing step S436, a past time section to where the scalar value of the current of banks other than the estimation target bank is about the same as the value at the current time tn is searched. The specific method is as described in the above-mentioned portions of FIGS. 7a, 7b, 7c, 7d, and 8.

Next, in processing step S437, a difference vector of the vector value of the current on the upstream side of the branch is created between the past time section to of the search result and the current time tn. The difference vector corresponds to ΔI _0tn in FIGS. 7c and 7d. Next, in processing step _S438 , ΔI _0tn is regarded as ΔI _1tn, and a declination candidate is created using a triangle consisting of I _1tn , I _1to , and ΔI _{1tn in} FIG. 7c or 7d. The reason why the same declination angle candidate is obtained is that all the interior angles are determined because the lengths of the three sides of the triangle are known. Furthermore, the deviation angle of ΔI _1tn is approximately equal to the deviation angle of ΔI _0tn . Further, the deviation angle of ΔI _0tn is known. This is because the time difference between the current vectors on the upstream side of the branch where the argument is known is ΔI _0tn . Next, in processing step S439, the value to be used as the argument value is narrowed down from the generally two argument angles using the method of narrowing down the argument candidates shown in FIG. These operations are returned to the processing step S434, and are repeated for all banks or banks that require estimation.

  In processing step S440, the deflection angle value of the current is converted into an amount that matches the purpose of use, if necessary, and is output as the estimation result of the power flow value and the power factor value.

  In the present invention described above, a method of estimating the power factor of the distribution system was proposed. The method of the present invention estimates the power factor on the lower side by using the power factor value on the upper side of the branch and the effective value of the current on the lower side of the branch, targeting the branch point in a distribution substation or the like. It is a thing. As a result of verification using the actual measurement data, it was confirmed that it is possible to estimate with high accuracy the average error for the three branch points.

102: Upper substation such as intermediate substation 103: Distribution substation 104: Transmission line 106: Large-scale consumer load 107: Small-scale consumer load 108: Transformer 110: Photovoltaic power generation Distributed power source 111 such as solar power: Transmission Upper side of distribution system 112: Lower side of transmission and distribution system 120: Outgoing measured value of upper substation 102: Incoming measured value of distribution substation 103 122: Lower measured value of branch 123: Lower branch of branch Side estimation target power flow value 313: measurement value acquisition unit 314: measurement value conversion 315: estimation target bank selection unit 316: past time to determination unit 317: set value input unit 318 such as threshold value: integrated control unit 321: measurement value Storage unit 322: time difference creation unit 323: declination candidate creation unit 324: declination candidate narrowing unit 325: conversion from declination to physical quantity of usage pattern 329: estimation result output

Claims (7)

The current considering the phase against the voltage is input and stored as a vector value from the upper side of the branch of the power system, and the current is input and stored as a scalar value from each branch line on the lower side of the branch,
Select one of the multiple branch lines on the lower side of the branch as the target of power flow estimation,
For a scalar value of current in branch lines other than the selected branch line, obtain a past time having a past value that is approximately equal to the current value,
For the vector value of the current on the upper side of the branch, obtain the difference vector of the vector value at the current time and the past time,
From the difference vector, and the scalar value of the current at the past time and the current time of the branch line of the power flow estimation target, obtain a candidate for the argument of the current value of the branch line of the power flow estimation target,
Narrow down the candidates for the angle of deviation using the statistical property of the angle of deviation of the current value,
A power flow value estimation method for an electric power system, which comprises obtaining a physical quantity corresponding to a power flow value from a value of a narrowed angle of a current of a branch line of a power flow estimation target. A power flow value estimation method for a power system according to claim 1,
The past time having a past value that is substantially equal to the current value is a time within a time range in which the load fluctuation of a customer of the power system can be regarded as substantially constant between the current time and the past time. A method for estimating power flow in a power system. It is a power flow value estimation method of the power system in which the current is measured as a vector value considering the phase with respect to the voltage on the upper side of the branch of the power system, and the current is measured as a scalar value on the lower side of the branch.
A first step of obtaining a current vector value on the upper side of the branch of the power system and a current scalar value on the branch line on the lower side of the branch;
A second step of selecting one of a plurality of branch lines on the lower side of the branch as a target of power flow estimation;
A third step of obtaining past time cross-section data that is substantially equal to the current value for the scalar value of the current in branch lines other than the selected branch line;
A fourth step of calculating a difference vector between the vector value of the current on the upper side of the branch and the vector value of the current on the upper side of the current branch in the past time cross section that is substantially equal to the current value;
A candidate for a declination of the current value of the branch line of the power flow estimation is obtained using the difference vector and a triangle composed of the past time section of the branch line of the power flow estimation target and the scalar value of the current 5 steps,
A sixth step of narrowing down the candidates for the angle of deviation using the statistical property of the angle of deviation of the current value,
A method for estimating a power flow value in a power system, comprising a seventh step of converting a value of an angle of deviation of a current of a branch line on a lower side of a narrowed branch into a required physical quantity such as a power factor value. The power flow value estimation method for a power system according to claim 3,
In the third step of obtaining the data of the past time section where the scalar value of the current other than the selected branch line is substantially equal to the current value, the range of the past data is set to 1 hour or 30 minutes. Method for estimating the tidal current value. The method for estimating a power flow value of a power system according to claim 4,
A power flow value estimation method for a power system, characterized in that a range of past data is set to a time at which a load variation of a customer connected to the power system is within 20%. The power flow value estimation method for a power system according to any one of claims 3 to 5,
After creating a histogram for the argument values of a plurality of current value argument candidates, the frequencies are added together, and the argument candidate closest to the mode of the histogram is determined to be the estimated value of the argument. Method for estimating power flow value of power system. A measurement value storage unit that inputs and stores a current in consideration of the phase with respect to voltage as a vector value from the upper side of the branch of the power system, and inputs and stores the current as a scalar value from each branch line on the lower side of the branch,
A declination estimation target bank selection unit that selects one of a plurality of branch lines on the lower side of the branch as a target of power flow estimation,
For a scalar value of current in branch lines other than the selected branch line, a past time point determination unit that obtains a past time having a past value that is substantially equal to the current value,
Regarding the vector value of the current on the upper side of the branch, a difference vector calculation unit that obtains a difference vector between the current time and the vector value at the past time,
From the difference vector, and the scalar value of the current at the past time and the current time of the branch line of the power flow estimation target, a candidate creation unit that obtains a candidate for the declination of the current value of the branch line of the power flow estimation target,
A declination candidate narrowing unit that narrows down the declination candidates using the statistical property of the declination of the current value,
A power flow value estimation device for a power system, comprising: a conversion unit that obtains a physical quantity corresponding to a power flow value from a value of a narrowed angle of a current of a branch line that is a power flow estimation target.

Images