JPH0226451B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention deals with ground faults in non-grounded or high-resistance grounding systems by forcibly grounding the grounding phase in order to reduce damage to grounded objects. The present invention relates to a method for determining the presence or absence of a substance at high speed.

Normally, in high-voltage power distribution systems, the neutral point of the system is ungrounded, and a GPT for zero-phase voltage detection is used to detect high-resistance ground faults such as electric shocks and contact with trees.
A resistor of several tens of ohms is simply inserted into the open delta circuit. In a 6.6Kv system, this resistance is usually 10KΩ on the high voltage side, so even if the resistance value of the grounded object is, for example, about 2KΩ,
The zero-sequence voltage Vo at that time is the ground capacitance Co of the system
is negligible, 83%Co of the normal earth voltage E
= 50% at 1μF/φ, Co=17% at 3μF/φ, Co=
At 5μF/φ, it becomes about 10% and detection is possible, but
Due to time limit coordination, the power supply side does not shut off the power immediately.
If a ground fault continues for about 1.5 seconds or more, the feeder is shut off.

Of course, it may disappear naturally during this time, but in the case of a personal accident due to electric shock, it cannot be saved, and as the number of cable systems has increased recently, secondary accidents may occur due to abnormal voltage generation due to intermittent ground faults. It may develop into.

One way to prevent this is to forcibly ground the faulty line at high speed, making it voltage-free so that no current flows to the original ground fault location. This can prevent electrocution, reduce damage caused by ground faults, and improve the probability of retransmission.

Various methods have been proposed for detecting ground faults in the past, but all of these methods are based on the assumption of vector values. There may be errors, and depending on the ground fault phase, the values corresponding to the healthy phase may be
There were cases where it took a long time to exceed a certain level.

This invention was made in view of the above points,
The object of the present invention is to solve the drawback of conventional ground fault phase detection devices, which is that it takes time to detect, and to provide a device that can detect ground faults at high speed.

The following describes an AC voltage that includes a transient phenomenon according to the present invention (or a case where the voltage is less than or equal to one current)
The instantaneous value v ₀ sampled at fixed time intervals td
(t=t ₀ ), v ₁ (t=t ₁ =t ₀ +td), v ₂ (t=t ₂ =t ₀ +
2td), v ₃ (t = t ₃ = t ₀ + 3td), AC voltage v = √
A method of determining the effective value v and time constant τ=1/α of 2V {sin(ωt+φ ₀ )−sin φ ₀ ·exp(−αt)} in steady state will be described.

First, the behavior of the fault voltage waveform (zero-sequence voltage) at the time of a ground fault accident in a high-resistance grounded or non-grounded system will be explained.

Figure 1 shows an ungrounded high-voltage distribution line, and it is assumed that a one-line ground fault has occurred due to resistance Rg.
In this case, a voltage for monitoring the potential v ₀ at the neutral point appears across R _N , but a demand arises to detect this much faster than when v ₀ reaches its maximum value, for example within several milliseconds.

FIG. 2 shows the three-wire zero-phase equivalent circuit of FIG. 1. Here, r/3 is the ground resistance of the three wires including the control resistance R _N (Fig. 1) in the leakage resistance of the line, and if the leakage resistance is ignored, r/3 = 1/9 × (zero-phase transformer Turns ratio) is expressed as ² × R _N.

In FIG. 2, sw is a switch that simulates the occurrence of a ground fault, and the voltage v ₀ when this sw is closed is expressed by the following equation with respect to the line voltage √2Esin (ωt+).

Here, δ is the phase of the line voltage when a ground fault occurs, and δ is the value expressed as δ=arctan3ωcR _g /(1+3R _g /r) (2), and is the electrical angle at which v ₀ lags the line voltage E. represents. The following relationship is obtained by comparing equations (8) and (1).

₀ = −δ ………(4) α=1+3R _g /r/3cR _g ………(5) From (3)(5) (1+3R _g /r)=3v ² /(1+ω ² /α ² )E ² ......
(6) Assuming that E and r are known, Equation (16),
By using the result of (17), R _g can be calculated from equation (6).

i.e. From the above, the zero-sequence voltage in the transient state after a ground fault occurs is v(t)=√2v{sin(ωt+ ₀ )−sin ₀・e
xp(−αt)}……(8) In the following, this v(t)
A method for calculating the effective value v of the zero-sequence voltage after the end of the transient state from the instantaneous value of is described below.

First, when v(t) is differentiated with respect to t, dv/dt=v'=√2v{ωcos(ωt+ ₀ )+αsi
n ₀・exp(−αt)}……(9) is obtained. Similarly, d ² v/dt ² = v″=√2V {−ω ² sin (ωt+ ₀ )
−α ² sin ₀・exp(−αt)}……(10) d ³ v/dt ³ =v=√2V{−ω ³ cos(ωt+ ₀ )
+α ³ sin ₀・exp(−αt)}……(11) is obtained. From equations (8) and (9), v′+αv=√2V{ωcos(ωt+ ₀ )+αsin(
ωt+ ₀ )}……(12) Here, if ψ=arctanω/α, then v′+αv=√2V√ ² + ² sin(ωt+ ₀ +ψ
)……(13) Similarly, v″＋αv′＝√2Vω√ ₂ + ₂ cos(ωt＋ ₀ ＋
ψ)……(14) −(v+αv″)=√2vω ² √ ² + ² sin(ω
t+ ₀ +ψ)……(15) From equations (13) and (15) α=−ω ² v′+v/ω ² v+v″……(16) From equations (13) and (14) α and V can be calculated from equations (16) and (17). However, the value of v′v″v to be given in equations (16) and (17) is the data v ₀ obtained from the circuit,
It must be expressed as v ₁ , v ₂ , and v ₃ . This can be obtained as follows by taking up to the third-order term of the Taylor expansion.

When v ₁ , v ₂ , and v ₃ are expanded by Taylor, v ₁ = v ₀ + v′ (t ₀ ) td + v″ (t ₀ )/2・td ² + v
(t ₀ )/6・td ³ ………(18) v ₂ = v ₀ + v′ (t ₀ ) (2td) + v″(t ₀ )/2・(2
td) ² + v (t ₀ ) / 6 (2td) ³ ...... (19) v ₃ = v ₀ + v' (t ₀ ) (3td) + v'' (t ₀ ) / 2・(3
td) ² +v(t ₀ )/6(3td) ³ ......(20).

v'=v'( _t0 )=3( _v1 - _v0 )-3/2( _v2 - _v0
)+1/3(v ₃ −v ₀ )/td……(21) v″=v″(t ₀ )=−5(v ₁ −v ₀ )+4(v ₂ −v ₀ )
−(v ₃ −v ₀ )/(td) ² ………(22) v=v(t ₀ )=3(v ₁ −v ₀ )−3(v ₂ −v ₁ )
+(v ₃ −v ₀ )/(td) ³ ......(23) is obtained. Using these values, equations (16) and (17)
By calculating α, V, therefore, τ, V can be calculated.

The effective value of the zero-sequence voltage calculated above is compared with the tap value (K Tap), and when V>K Tap, it is considered a ground fault in the system.

During actual calculation, the zero-phase voltage is left as (V) ² to shorten the calculation time, and compared with (K Tap) ² . That is, ω ² (v′+αv) ² +(v″+αv′) ² /2ω ² (α ² +ω
² )＞(K Tap) ² ......(24) Next, we will explain the case where a digital processing device (for example, a microprocessor) is used as an example of the configuration of a device that implements the above method, based on Figure 3. do.

In Fig. 3, the voltages of the four phases A, B, C, and zero phase (in principle, only zero phase is sufficient, but for data checking etc., four phase input is used in this explanation).
is input to the device by a voltage sensor 31 and is converted to a suitable magnitude as input to a processing device 33 by an input converter 32. Normally, set the maximum input value to be 20 volts from peak to peak. The processing device 33 basically consists of an analog-to-digital conversion section (A/D conversion section) 34, a high-speed data transfer device 35, an arithmetic processing section 36, and an output section 37, and executes a series of procedures of the present invention. do.
The four-phase voltage from the input converter 32 is transferred to the A/D converter 3.
4, and the sample and hold circuit 34
1, the signal from the control circuit 345 is sampled at certain fixed time intervals. This value is converted into a digital quantity by an analog-to-digital converter 343, but in this example, since the analog-to-digital converter 343 is shared for a plurality of inputs, a multiplexer 342 is placed at the front stage. The digitized input is a buffer circuit 344
can be stored in The digital amount stored in the buffer circuit 344 is input to the data memory area 362 of the arithmetic processing section 36 by the data transfer device 35.

In the arithmetic processing unit 36, the program memory 36
3, the processing shown in FIG. 4 is executed.

That is, first input the data into the data memory,
After checking the input data, if there is no abnormality in the data, the calculations of equations (21), (22), and (23) are performed. Thereafter, the presence or absence of an accident is determined using equation (24). The setting value (K Tap) ² used at this time is stored in the setting memory.

If the conditional expression (24) continues for a certain period of time, it is assumed that a ground fault has occurred and a control signal is output. This certain period of time is redundant to avoid outputting an erroneous output due to a single data error, so it can be determined from the overall completion time.

This output signal is outputted from the output section 37 in FIG. 3 and outputted to the equipment side (for example, a grounding mechanism) through the output interface 38, thereby performing a forced grounding operation or the like.

Next, in order to check the finishing time of the device according to the present invention,
Explain the time chart.

When sampling is performed at time intervals td as shown in FIG. 5, A/D conversion and data processing must each be completed at least within td. Therefore, the data sampled at the worst time tn will be processed within the td time starting from tn+1. In other words, when considering finishing time, it is only necessary to consider the worst-case TD time delay.

Next is the sampling timing and accident occurrence timing, which also have variations in the maximum tn. That is, if an accident occurs immediately after sampling tn, accident data will only be input at sampling tn+1. This also causes a td time delay. From the above, the shortest completion time for this device is 6 td, if the above-mentioned margin (redundancy) is not taken into consideration, since four pieces of time-series data are used and the worst time delay is 2 td. If the sampling frequency is 1ms (i.e. td=1ms), then 6
Accidents can be detected in ms. Here, td is determined by the computing power of the digital processing device.

As described above, the present invention is a ground fault detection device that is characterized by being able to detect ground fault accidents at extremely high speed and take appropriate measures.

[Brief explanation of drawings]

Fig. 1 is a 3-wire diagram at the time of a ground fault accident in an ungrounded high-voltage distribution line, Fig. 2 is a 3-wire collective zero-phase equivalent circuit diagram, and Fig. 3 is a configuration of a ground fault detection device according to the present invention. A block diagram showing an example, FIG. 4 is a schematic processing flow diagram of this device, and FIG. 5 is a processing time diagram.
It's a chat. In the figure, 32 is an input converter, 34 is an analog-to-digital converter, and 36 is an arithmetic processing unit.

Claims (1)

[Claims] 1. From the zero-phase instantaneous value voltage that includes a transient phenomenon at the time of a high-resistance or ungrounded power system fault, the first derivative value v', the second derivative value v'', and the third Calculate the differential value v The voltage at the time of ground short circuit occurrence v = √2V {sin (ωt + ₀ ) − sin ₀・e ^- 〓 ^t } and the effective value v of the transient term sin ₀・e ^- 〓 ^t at the time of attenuation The constant 1/α is detected faster than the transient term of v disappears, and furthermore, without waiting for v to reach the peak value or pass the zero point, and its value is compared with a constant value proportional to the ground voltage E under normal conditions. A ground fault detection device characterized in that it detects a ground fault accident by causing a major event.