JPH02119528A - Differential relay system - Google Patents

Abstract

PURPOSE:To improve identification ability for a relay whether it is exciting rush current or accident current by allowing the operation of a main decision elements only when the differential current flows in both the positive and nega tive directions or when there is a sufficient width in a conduction period. CONSTITUTION:Let the set value t3 of a width decision element 5 be e.g., 260 deg. at an electric angle. When the pallet width PW is at the limit value, the nega tive current peak value hn is 0.35. In order to operate a bidirectional element 4 overlappingly so as to cause no blind spots, an allowance is made to the set value k1 of the bidirectional element 4 against this 0.35, letting it be the amplitude value of required operating current X0.30, etc. To no purpose this set value is highly sensitive to the required operating current and is reasonable enough. If the accident current is at not less than the set value in the width PW, the width decision element 5 produces outputs. If it is at not more than the set value, the current in the negative direction gets to not less than the set value k1 and the bidirectional element 4 operates for sure. This is the same in case the relation between positive and negative directions is in the other way. In exciting rush current, there is no bidirectional conduction and the width PW is so smaller than the set value t3 that no elements will produce outputs.

Description

[Detailed description of the invention] [Object of the invention] (Industrial field of application) The present invention relates to a calculation method applied to a differential relay (prior art) Technical aspects of a conventional differential relay for protecting a transformer The challenge is to distinguish between fault current and magnetizing inrush current.

As is well known, differential relays have a high ability to identify internal and external faults because they obtain the amount of operation from the sum of currents at each terminal. However, in the case of a differential relay for transformer protection, even though there is no accident if there is an eJ magnetic inrush current, a sum of currents called a so-called difference current (hereinafter referred to as difference current) appears and the same behavior as an internal accident occurs. produce quantity. For this reason, conventionally, a method has been used to distinguish between an internal fault current and an l1ilJ magnetic inrush current, for example, based on the content rate of the second harmonic. That is, the excitation inrush current is generated by the magnetic saturation of the iron core, but one cycle has a saturation period and a non-saturation period, and no current flows during the non-saturation period. Such a waveform has a large content of second harmonics, and on the other hand,
The fault current contains almost no second wave. Conventionally, this principle has been used for identification, and a typical boundary value for identification has been about 15%.

(Problems to be Solved by the Invention) As is well known, in recent years, as the charging capacity of the grid has increased, the content of low-order harmonics in the fault current has increased, resulting in insufficient identification ability. Furthermore, there may be cases where the above boundary value is exceeded.2. This identification agent is no longer viable.

The present invention has been made in view of the above circumstances, and it is an object of the present invention to provide a method for distinguishing between excitation inrush current and fault current without depending on the content rate of the second investigation.

[Structure of the Invention] (Means for Solving the Problems) The main purpose of the present invention is a waveform discrimination method that directly discriminates the waveform of a difference current. That is, when the differential current flows in both positive and negative directions, or when the width of the energization period is sufficient, it is determined that it is a fault current, and the operation of the main determination element by, for example, a well-known ratio differential element is allowed, and both of the above conditions are met. If neither of these holds true, the operation of the main determining element is blocked.

(Production...) In the case of an accident, so-called differential current flows in both positive and negative directions, or the width of the energization period exceeds a certain value. On the other hand, in the case of the l1JJ magnetic inrush current, it flows only in one direction, and the non-energizing period is greater than a certain width, and the width of the energizing period is less than a certain value. Therefore, the above means allows the main determining element to operate only at fault current.

(Embodiment) FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention. In the figure, 1 is a main determination element, which is made of, for example, a well-known ratio differential element. 2 is a waveform discrimination element that discriminates the waveform of the difference current to permit or block the operation of the main judgment element 1;

The details of the ratio differential element will be omitted since it is a well-known technique, but a plurality of currents 11 to 13 are input, a so-called difference current 1d is derived, and ratio differential determination is performed to generate an output OP.

Here, the currents 11 to 13 are, for example, the primary to tertiary currents of a three-winding transformer, and the conversion of the current transformation ratio is performed using a well-known method.The number of turns of the ct transformer is 2. p) Is Rui generally N'? ：It goes without saying that it is the same even if there is.

The waveform discrimination element 2 includes a bidirectional element 4, a width determination index 5, and an OR
The bidirectional element 4 detects that the differential current flows in both positive and negative directions upon receiving the differential current 1d, and determines whether the width of the energization period is greater than a certain value by determining the width. , if either one of them holds true, OR
The I function of the five-way bidirectional search 4 and the width determination search 5 that produce the allowable output TA by the element 6 will be described below.

FIG. 2 is a block diagram illustrating an example of how the bidirectional element 4 is constructed. 7 is a level detection element that detects a positive wave.

8 is a time element. When the current 1d exceeds the detection level, element 7 produces an output, which is applied to element 8. Element 8 produces an output when the input from element 7 continues for a certain period of time C1 or more, and even if the input disappears, it maintains the output for a certain period of time t2. A certain period of time C2 is a minute period of time to confirm that the input exceeds the detection level of 1.
is for adjusting the deviation from the judgment time of the negative wave and the deviation from the judgment time of the main judgment element. That is, the determination times of each element do not completely match and deviate even within one cycle, and in order for the AND to be established, it is necessary to extend the time a little and wait for other signals. Element 9 produces an output in place of the positive wave of element 7 when the magnitude of the negative wave, i.e. -1,i, exceeds 1 at the detection level and is applied to element 10.

Element 10 has the same functionality as element 8. element 8 and element 1
An output of 0, ie, both inputs of the A and N D elements 11, will produce the permissible output PN. As mentioned above, since the current flowing in both directions is not the wJ magnetic inrush current, one of the intended purposes can be achieved with this.

FIG. 3 is a block diagram showing an example of an ellipse of the width determination summary 5. 12 is a level detection element, and 13 is a time element. When the absolute value of the difference current 1d exceeds the detection level by 2, element 12 produces an output that is applied to element 13. element 13 is element 12
If the input from 63 continues for a certain period of time or more, width judgment output 1
40, and the output is 140 for a certain period of time t4 after the input disappears.
is maintained. The constant time t3 is a limit value for determining the width of the energization period, and the constant time t4 is for adjusting the deviation from the main determination element time of the detached house according to element 8 or 10. This width determination principle is effective in cases where the bidirectional element does not produce an allowable output due to a fault current. In other words, when the well-known DC component current determined by the accident occurrence phase is large,
Although the fault current is offset to one side, in the case where the bidirectional element does not operate, the width of the energization period is sufficiently wide and the width determination point produces the output WD.

FIG. 4 is a waveform diagram of a typical example of the magnetizing inrush current, which is a prerequisite for explaining the effects of FIGS. 2 and 3. That is, for example, a positive current flows during the period pw.

It does not flow during the remaining pz period. 15% of the second harmonic
pw corresponds to approximately 246 degrees in electrical angle.

FIG. 5 is a waveform chart for explaining the effects of FIGS. 2 and 3. This figure represents the relationship between the width OW of the current conduction period in the positive direction and the current peak value hn in the negative direction when the sine wave is offset by a DC component.

That is, when pw is 290° to 250° in electrical angle, hn is 0.18 to 0.43, where the amplitude of the sine wave is 1.

It will be explained below that with such a relationship, the scope of assignment between the bidirectional element in FIG. 2 and the width determination summary in FIG. 3 can be easily matched. The setting value C3 of the width judgment summary is set to the above 2 in electrical angle.
For example, if we take the allowance into account for 46° and set it to 260°, then! When pw is the limit value, the negative current peak value hn is 0.3
It is 5. In order for the bidirectional elements to operate in an overlapping manner without blind spots, the setting @ of the bidirectional elements is set to 1 with a margin of 0.35, and the required operating current amplitude value xo is set to 30, etc. This setting value is a reasonable value without making the sensitivity too high for the required operating current. If the fault current causes the width pw to be greater than or equal to the set value, the width determination index produces an output, and if it is less than that, the current in the negative direction is greater than or equal to the set value, and the bidirectional element operates reliably. The same applies even if the positive/negative relationship is reversed. With the excitation inrush current, there is no bidirectional energization, and the width pw is also sufficiently larger than the set value 63, so none of the elements produces an output.

For the sake of simplicity, the case where there is no distortion in the fault current has been described above, but there is no contradiction in the above explanation even if the second harmonic, which has the most influence on this determination, is considered. First, even if there is a second harmonic in the fault current, the waveform shown in FIG. 4 will not occur. The reason for this is that, as disclosed in Japanese Patent Application No. 57-23022, even if there is a second harmonic, the phases of the fundamental wave and the second harmonic of the fault current are approximately the same, and such phase Including in rA section *
Regardless of the rate, a waveform in which the current is close to 0 over a wide range does not occur. Furthermore, although the relationship shown in FIG. 5 changes slightly depending on the distortion, no particular problem will occur if the set value has a margin of 1.

As described above, according to the above embodiment, the characteristics of the waveforms of the fault current and the vjJr:a inrush current can be determined using simple means such as the level detection element and the time element, regardless of the second sinus wave content rate. By directly detecting and identifying the bidirectionality or the width of the energization period, it is possible to avoid the intended purpose.

FIG. 6 is a block diagram showing the configuration of another embodiment of FIG. 2. , the second embodiment has a level detection element 15 in contrast to FIG.
and a time element 16 are added, and the AND element 11 is replaced with an AND summary element powered by three people. The level detection element 15 has the absolute value Ia l of the difference current 1d as the detection level'
exceeds 3, an output is generated and applied to time element 16. The time element 16 is the same element as 8 or 10. When all of these outputs are satisfied and applied to AND element 14, a permissible output PN is produced.

In this embodiment, the detection level 1 is the amplitude of the required operating current (a fraction of the singing value) as described above, so 1. Suitable for checking just in case.

FIG. 7 is a block diagram showing the configuration of another embodiment of FIG. 6. 7, 8, 9 and 10 are the same as in Fig. 6;
level detection elements 17 and 18 in place of 15 in the figure;
16 in the figure is replaced with one time element and 20.

The level detection element 17 generates an output when the magnitude of the negative wave of the difference current 1a, that is, the magnitude of the negative wave of the difference current 11d, exceeds the detection level 3. The level detection element 17 performs confirmation and time adjustment using a time element 19 similar to the time element 8, and performs an output from an AND element 21G. AN with the output of time element 8
Take D. The level detection element 18 produces an output when the difference current 1d exceeds the detection level by an AND element 22 through a time element 20 similar to the time element 19.
AND with the output of . 23 is an OR element, and if there is an output from either the AND element 21 or 22, the allowable output PM
will occur. The action in FIG. 7 is equivalent to that in FIG. 6, and various equivalent modifications are possible without changing the gist of the present invention.

FIG. 8 is a block diagram showing the configuration of another embodiment of the width determination guide 5. In FIG. This embodiment is the same as that shown in FIG. 3 except that a time element 24 is added to the width determination summary shown in FIG. The time element 24 receives the output of the level detection element 12, produces an output without delay, excluding a minute time required for processing (the same applies hereinafter), and maintains the output for the time t5 after the input disappears. In other words, this width determination summary means that even if the output of the level detection element 12 is interrupted for less than time t5, the output is considered to continue. I will explain next time that the effect is the same even with this method.

FIG. 9 is a waveform diagram for explaining the effect of FIG. 8.

1d is an example of the extremely distorted differential current waveform, and dl is its absolute value. Although the output of the level detection element 12 disappears even when the absolute fi I Idl becomes less than the set value 2 or during a period such as t7, the output of the time element 24 continues. In the excitation inrush current, as described above, there is a non-energizing period of a certain value or more, so time t5 is set to a value smaller than that. Further, the set value C3 of the time element 12 is set to a value larger than that in the case of FIG. 3 by t5. If you do this, the waveform will be extremely distorted and 1. Even if a short time is over, the input of the time element 13 continues to make a correct judgment. In the case of excitation inrush current, the output of the time element 24 is longer than the width of the energization period by a time C5, but since the set value C3 of the time element 13 is made longer by that value, there is no error in judgment.

The above embodiments can be implemented not only for analog input but also for sample value input. In particular, the bidirectional elements in FIG. 2, FIG. 6, or FIG. 7 do not require any additional explanation. Regarding the width judgment summary in Figure 3 or Figure 6,
This can be achieved as is if the input sampling density is sufficient. However, in the so-called digital relay, sample values at intervals of, for example, 30' electrical angle are used due to calculation burden and the like.

An example in such a case will be described below.

FIG. 10 is a flowchart showing the flow of another embodiment of the width determination summary shown in FIG. This is an algorithm for making judgments.

Difference current 1. This calculation is started every time each sample value of i arrives, and in step S1 it is determined whether the set value is 2 or more, and Y
If so, proceed to step S2; if N, proceed to step S3.

In step S2, it is determined whether the determination Y of step S1 is 7 consecutive samplings, and if Y, the process advances to step S5, and if N, the process advances to step S8. Step S3 and step S4
correspond to step S1 and step S2, respectively.

That is, the determination is made regarding a negative wave instead of a positive wave, and the process is the same including the destination after the determination, except that in step S1 the determination is made based on Ia>kz, whereas in step S3 the determination is made based on -1a>kz.

Step S5 is a step for approximately calculating the width of the energizing period of the positive wave or the negative wave. A sample value of the difference current 1d at one point in time is represented by - (the subscript d is omitted to avoid complexity). Using the latest value 1m, the value 3 times ago 1m-3, and the value 6 times ago +71-6, the approximate value of the width is calculated by the following formula. That is, the width pw expressed in electrical angle is pw・180' +
, 30° (1m-a/(i!1-6-A)+j@/
(iffi-B) )・・・・・・(1) However, A=3(In+In-5)/4-i! ! i-3
/2-J 3jil -11-6)/4B=3(!a
+1111-6)/4-1g-3/2 + i 3
It is determined by fi@-j color-6)/4. Step S6
determines whether the above-mentioned width pw is equal to or greater than a certain value 63, and if Y, the process proceeds to step S7 to generate the allowable output WD and 1. If N, proceed to Step S8.

Step S8 is a so-called off-delay element, which delays the output WD by a certain period of time t4, and then ends the process and waits for the next sample.

FIG. 11 is a waveform diagram for explaining the effect of FIG. 10. In step S2 or S4, the positive wave or negative wave is 7.
Since the sample continues and reaches step S5 with the detection level exceeding 2, the latest sample value I11
．． The value 1m-3 three times before and the value IM-6 of 6 continue to be positive or negative values. Although the figure shows a positive wave, the following explanation is exactly the same for a negative wave.

The meaning of the calculation method is as follows. That is, sample value 1r
r,! Apply m-= and 1n-6 to a sine wave with DC offset to find the DC component and sine distribution, and then sample the waveform @I#! as well as! Assume a value extending 30 degrees on both sides of a-e. The illustrated points Ca+÷1 and CIl
+-7 are these assumed values. Note that these assumed values are represented by A and B, respectively, in step S5 of FIG.

Next, the assumed value Cm-y, the sample value 1n-6, and the assumed value C
For each combination of m+1 and sample value IIn, the zero passing point and sample value In-6 or 1 are determined by linear approximation.
#Calculate 1mW2 or W3 with I. In addition, the width p
The interval between sample value 1 corner of w and 1 m-6 is 30 as mentioned above.
``Due to the premise of interval sampling, it is 180°.

The value at time n of the waveform to be applied is n= D+PcoS((n-rg+3)30°)+
Qsin((n-n+3)30°)...(2
), then the following formula holds true.

For n = n-6, 11+-6= D -Q
−”...(3) n = ll-3, 1m-3□
D +P −・”(4) n=11 then I
n □ D +Q --(5)(3)~(5
), P and Q are determined from equation (2).In equation (2), the so-called DC component is assumed to be a pure DC component without attenuation, but compared to the time constant of the magnetizing inrush current, the time of one period of the fundamental wave is It is sufficiently short that it is a practical approximation.

Next, find the assumed value Cn from equation (2) with n = lIzu and n = n+1. If the intervals between the assumed values CJ+-7 and CJl+1 and the adjacent zero passing points are approximately W1 and Wq, then linear approximation is obtained. Therefore, the following formula holds true.

By the way, the width pw is l"=18"W2" W3, and based on the premise, Wl"W2" W3+Wq ・30
Therefore, the above equation (1) is obtained using equation (8).

According to this embodiment, it is possible to calculate the width of the energization period by capturing the waveform in a wide range, and it is possible to avoid errors in the rising portion of the current and make a correct determination.

In this example, approximate calculations were performed using sample values at 90° intervals in electrical angle, but in general, if there are three sample values, not limited to 90° intervals, the DC component and the sine distribution can be obtained. The width of the energization period can be approximately calculated using a similar method.

FIG. 12 is a flowchart showing the flow of another embodiment of FIG. 10. This example is similar to FIG. 10 except that the approximate calculation is performed using three sample values at 60° intervals
Step S2 in Figure 0. Steps S9. instead of S4 and S5 respectively. Except for the replacement in SIO and S11, it is a rotation.

In contrast to the case of FIG. 10 in which 7 consecutive samplings are confirmed, in steps S9 and S10 of this embodiment, 5 consecutive samplings are confirmed. This is because three sample values spaced at 60' intervals are used. In other words, the sample values at 30° intervals are 5
Time 3! This is because the approximate calculation becomes effective only when the value subsequently becomes positive or negative.

Step S11 is a step of approximately calculating the width pw of the energization period. The meaning of the calculation method and the calculation content will be explained together.

From the sample values 1m1n-2 and 1m-4, calculate the DC component and sine distribution in the same way as in Figure 10, and then extend the waveform by 30 degrees on both sides of the sampled values 1m and 1m-4 to obtain the values Crt+1 and Assume CJI-5. However, the assumed value of : is expressed as approximately G and 11. Expected value G=
(1-5, sample value 1m-4 and assumed value H=Cn÷1
By linear approximation for each combination of sample value m and
5 or W6. In addition, a sample of the width pw @
The distance between 1 m and 1 m-4 is 120° based on the above-mentioned 30° sampling assumption.

The value at time n of the waveform to be applied is 1/l=D+Pcos(in-n+2)30°)
+Qsin((n-n+2)30° (9) The following equation holds true.

For n = tg-4, lll-4 = D + P/2
-r3G/2...(10) When n = m-2, +71-2 = D + P
...---(11) When n=1m, 1n = D, P/2+, r3Q/2 ......(12) When 0 and Q are found from equations (10) to (12), D =
IJllm-4-! J! −2,Q =(In −1n−
4) #'3...(13) Next, assuming n = n-5 and n = 1÷1, calculate the expected value Cn from equation (9), G = (@-5 = D-Q, H=(:!n44 =D
+Q (14) If the distances between the assumed values Cm-5 and CJl+1 and their adjacent zero passing points are respectively w5 and W6, then the following equation holds true by linear approximation.

G= Cl-5=D-Q=Is Im-+ -IM-2-(IIR-I
JI−+)/(3−(16)H”Ctn+1 = D
+ Q□ IjIla-t -1111-2+
(I!II-IJI-4)/J3-(17)pw=12
0° + νν”5 ”Ws=120' + 3
0° (IJI-4/ (IIR-4-G)+ In,
3 The other steps are the same as those shown in FIG. 10 as described above, and the effect is almost the same as in FIG.

Although the embodiment of FIG. 10 or FIG. 12 uses exactly three sample values necessary to determine the parameters, it is possible to use more than three sample values and calculate the DC component and the sinusoidal component using the well-known least squares method. It is also possible to calculate the minutes.

Further, although FIG. 11 shows that the number of sample values of the difference current i exceeding the detection level of 2 is exactly seven, it is not necessary that the number is exactly seven, and any number greater than that is sufficient. If there are more than 7, that is, multiple combinations of 3 sample values spaced at 90° intervals, calculate all combinations for confirmation, or use the combination containing the largest sample value for the most probable approximation. Various changes can be made without changing the gist of the present invention.

In the embodiment shown in FIG. 1 described above, the output of the waveform discriminating element is the allowable output TA, but it can also be easily considered that the output is the blocking output. Although not particularly shown in the figure, it is logically equivalent to create a blocking output when the differential current does not flow in both directions and the width of the energization period is smaller than a predetermined value, and the same effect as in FIG. 1 can be achieved. It is clear that

[Effects of the Invention] As explained above, according to the present invention, the characteristics of the waveforms of the fault current and the excitation inrush current are directly captured and identified by the bidirectionality or the width of the energization period. It is possible to provide a differential relay system capable of detecting a fault current regardless of the content rate of second harmonics.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention, FIGS. 2 and 3 are block diagrams showing the internal configuration of FIG. 1, and FIGS. 4 and 5 are block diagrams showing the internal configuration of FIG. 3 is a waveform diagram for explaining the operation, FIG. 6 is a block diagram showing the configuration of another embodiment of FIG. 2, FIG. 7 is a block diagram showing the configuration of another embodiment of FIG. 6, 8 is a block diagram showing the configuration of another embodiment of FIG. 3, FIG. 9 is a waveform diagram for explaining the operation of FIG. 8, and FIG. 10 is a flowchart of another embodiment of FIG. 3. FIG. 11 is a waveform diagram for explaining the action of FIG. 10, and FIG. 12 is a flowchart representing the flow of another embodiment of FIG. 1... Main judgment element 2... Waveform discrimination element 3...
・AND element 4...Bidirectional element 5...Width judgment summary 6...OR element 7...Level detection element
8... Time element 9... Level detection element 10...
Time element 11...AND summary element 12...Level detection element 13...Time element 14...AND
Element 15... Level detection element 16... Time element 1
7, 18...Level detection element

Claims (1)

[Claims] In a differential relay system comprising a main determination element whose operation amount is the sum of a plurality of currents, and a waveform discrimination element that discriminates the waveform of the operation amount and allows or blocks the operation of the main determination element, the waveform discrimination The element includes a bidirectional element that detects that the amount of operation is greater than a predetermined value when the polarity is positive and negative, and a width judgment element that detects that the width of the energization period of the amount of operation is greater than or equal to a predetermined value. , wherein the waveform discrimination element allows the main judgment element to operate when either or both of the bidirectional element or the width judgment element produces a predetermined output. Relay method.