JPH0670447A - Ground-fault detection method - Google Patents

Abstract

PURPOSE:To surely detect a momentary short circuit as a sign of a ground fault. CONSTITUTION:Whether the effective value of zero-phase voltage Vo exceeds a certain threshold value or not is a zero-phase voltage-determining condition, whether the effective value of the high-frequency component of ground-phase voltage VA, VB, VC exceeds a certain threshold value or not is a kick phenomenon-determining condition, and a state where both conditions are satisfied is detected as a sign of ground fault.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a ground fault detection method for detecting an instantaneous ground fault phenomenon or a fine ground fault phenomenon before a continuous accident.

[0002]

2. Description of the Related Art Conventionally, as a means for detecting an instantaneous ground fault phenomenon in, for example, a fine ground fault relay, there is a means for detecting an effective value level of a zero-phase voltage Vo. Further, even in a ground fault protection relay that protects the system from a complete ground fault, the ground fault accident determination is performed by combining the effective value of the zero phase voltage Vo with other information such as a zero phase current.

[0003]

The conventional ground fault protection relay has a zero-phase voltage V regardless of the analog or digital system.
Although the ground fault is determined based on the effective value level of o, a so-called ghost voltage may be superimposed on Vo due to a cause other than the ground fault, so that it is erroneously determined as a ground fault. There is. In addition, since the ground fault protection relay originally cuts off only the line that is operationally impaired due to the expansion of the accident when a ground fault is detected, it is possible to eliminate the accident point spontaneously such as an instantaneous ground fault. Must not break the line. For that reason, generally, a certain period of operation is provided. Further, a low-pass filter is generally provided to prevent malfunction due to noise components superimposed on the zero-phase voltage. Therefore, in the conventional ground fault protective relay, the waveform of the zero-phase voltage Vo is several cycles (0.1
.About.0.3 S) and is effective only when the effective value level exceeds a certain threshold value. Therefore, in the case where the accident phase is self-recovered within less than 1/4 cycle of the fundamental frequency like an instantaneous ground fault, the zero phase voltage V
Instantaneous ground fault cannot be detected only by o. As described above, the conventional ground fault protection relay is used only for protecting the ground fault as a continuous accident, and does not obtain maintenance information before the continuous accident.

If, for example, the insulator is cracked by several lightning strikes and water enters, an instantaneous ground fault or a fine ground fault may occur at that location. On the contrary, such a ground fault is continuously generated. The need for maintenance and inspection can be known by detecting it as a precursory phenomenon to the accident.

An object of the present invention is to provide a ground fault detection method capable of surely detecting an instantaneous ground fault phenomenon or a fine ground fault phenomenon as a precursor phenomenon leading to a continuous ground fault accident.

[0006]

In order to achieve the above object, the present invention obtains an effective value of a zero-phase voltage and determines whether or not the value exceeds a given threshold value. And a kick phenomenon determination process for determining the effective value of the high frequency component of each phase voltage and determining whether the effective value exceeds a given threshold value. A state in which both are satisfied is detected as a ground fault occurrence state.

Further, the difference value of the zero-phase voltage with respect to the value of the fundamental wave frequency before the integral period is obtained, the starting point of the ground fault is obtained from the change of the difference value with the passage of time, and the starting point is used as a reference. The kick time presence / absence determination time range is limited.

[0008]

According to the ground fault detection method of the present invention, first, the effective value of the zero-phase voltage is obtained, and it is determined whether or not the value exceeds a given threshold value, that is, whether or not the zero-phase voltage is generated. Further, the effective value of the high frequency component of each phase voltage is obtained, and it is determined whether or not the effective value exceeds a given threshold value, that is, the presence or absence of the kick phenomenon. If the zero-phase voltage is generated and a kick phenomenon is also generated, the state is detected as a ground fault occurrence state. Therefore, the kick phenomenon does not occur in the phase voltage, and an abnormal so-called ghost voltage generated only in the zero phase voltage is not erroneously detected. Further, since the effective value of the high frequency component of each phase voltage is obtained and the kick phenomenon is determined by whether the effective value exceeds a given threshold value, only the kick phenomenon caused by the ground fault is detected. Therefore, it is possible to reliably detect the occurrence of an instantaneous ground fault such that the accident phase self-recovers in, for example, 1/4 cycle shorter than one cycle of the fundamental wave frequency.

Further, the difference value of the zero-phase voltage with respect to the value of the fundamental wave frequency before the integral period is obtained, and the change point of this difference value with the lapse of time, that is, when the absolute value of the difference value suddenly increases. Obtained as the starting point of the connection. Then, the kick phenomenon presence / absence determination time range is limited on the basis of this starting point, and it is determined whether or not the kick phenomenon occurs at the time of starting. This makes it possible to accurately detect the kick phenomenon at the same timing as the start of the zero-phase voltage.

[0010]

1 is a block diagram of a ground fault detecting device according to an embodiment of the present invention. In FIG. 1, sample and hold circuits 1 to 5 sample and hold the phase voltage signals VA, VB and VC, the zero phase voltage signal Vo and the zero phase current signal Io, respectively. The multiplexer 6 selects one of the sample hold circuits 1 to 5 according to the selection signal and outputs it to the AD converter 7. The AD converter 7 converts the voltage signal into digital data. The CPU 8 executes a program written in advance in the ROM 9 to perform processing for ground fault detection. The RAM 10 is used as a working area for various calculations. The operation panel 12 sets various calculation parameters and displays the ground fault detection result via the operation panel interface 11. The printer 14 records and saves the detection result of a ground fault accident via the printer interface 13. Communication interface 1
Reference numeral 5 transmits the ground fault detection result and the like to another monitoring device via a communication line or a distribution line.

Next, FIG. 2 shows a processing procedure of the CPU 8 shown in FIG. 1 as a flowchart. Further, waveform examples of each signal are shown in FIGS. The process shown in FIG. 2 will be described below with reference to FIGS.

First, FIG. 3 is a waveform diagram of each signal before and after the occurrence of an instantaneous ground fault, where VA, VB and VC are phase voltage signals and Vo is a zero phase voltage signal. In this example, a ground fault occurs in the B phase at time to, a kick phenomenon occurs in any of VA, VB, and VC, and a voltage due to the ground fault occurs in the zero-phase voltage signal Vo.

Now, in step n1 shown in FIG. 2, the zero-phase voltage data V that has already been sampled and stored.
The value o is read in time series and the effective value is obtained.
This effective value calculation formula is as follows.

[0014]

[Equation 1]

Here, N is the number of data in the range for obtaining the effective value. For example, when sampling one cycle of the fundamental wave frequency at 256 points, N is set to 64 points corresponding to 1/4 cycle. The calculation of the equation 1 is obtained by the movement calculation for each data.

In step n2 of FIG. 2, it is determined whether the obtained effective value of Vo exceeds a predetermined threshold value. If the threshold value is not exceeded, it is considered that no ground fault has occurred, and the effective value calculation is performed on 64 points of Vo data including the data after one point (n2 → n1). If it exceeds the threshold value, the difference filter operation is performed on the zero-phase voltage signal Vo at n3. This is a process for removing the stationary fundamental wave component and the direct current component included in the Vo signal.
The data before the cycle is subtracted. This calculation is expressed by the following equation.

[0017]

## EQU00002 ## Vo ^s (i) = Vo (i) -Vo (i-256) FIG. 4 is a diagram showing the effect of the differential filter operation, where Vo is the original zero-phase voltage signal before differential filtering, Vo ^s Is the zero-phase voltage signal after differential filtering. In this way, the steady fundamental wave component and the direct current component are removed, and a zero-phase voltage signal containing only a pulsed signal due to a ground fault is obtained.

Subsequently, in step n4 shown in FIG. 2, the starting point of the ground fault is detected from the zero-phase voltage signal after the differential filtering. Specifically, it is determined whether or not the condition that the change amount of the current sampling data with respect to the previous sampling data exceeds a predetermined value Δv continues twice. For example, as shown in FIG. 5, the variation between the i-th data Vo ^s (i) and the next sampling data Vo ^s (i + 1) is ΔV.
And the next sampling data Vo ^s (i +
When the change from 2) also exceeds ΔV in the same direction, the timing of i is detected as the starting point.

Subsequently, in step n5 of FIG. 2, digital high-pass filter calculation is performed on each phase voltage signal VA, VB, VC. For example, FIR digital high-pass filtering is performed by the calculation of the following equation.

[0020]

[Equation 3]

Here, a (Z) has a cutoff frequency of 60.
A value that is 0 Hz is used. VN corresponds to any of VA, VB, and VC.

In the above digital high-pass filtering, if i is the starting point, VN ^h (i-512) to VN ^h (i-512) to VN
^h (i + 256) range, VN (i
It is performed in the range of −542) to VN (i + 256)).

Subsequently, in step n6 of FIG. 2, differential filtering is further performed on each phase voltage signal after high-pass filtering. As a result, the fundamental wave component slightly contained in each phase voltage after high-pass filtering is almost completely removed. As in the case of Equation 2, the calculation method obtains the difference from the data 256 points before, which corresponds to the data one cycle before the fundamental wave. Range if the calculation range at this time is the i and starting point ^{VN s (i-256) ~VN} s (i + 256), which if you look at the high-pass filtered data ^{VN h (i-512) ~VN} h ( i + 256).

FIG. 6 is a diagram showing the effects of digital high-pass filtering and differential filtering on the phase voltage signal. Here, as an example, the phase voltage VB of phase B, the data VB ^h after the high-pass filtering and further differential filtering are performed. The data VB ^hs shown in FIG. In this way, the kick component included in the phase voltage signal is extracted by the high-pass filtering, and the steady fundamental wave component is substantially completely removed by the differential filtering.

Subsequently, in step n7 of FIG. 2, an effective value is obtained for the data on which the digital high-pass filtering and the differential filtering have been performed on each phase voltage signal VA, VB, VC. In the same manner as Equation 1, the square root of the sum of squares of each digital is obtained for a certain range.
At this time, the effective value calculation range is set to a quarter cycle before and after including the starting point based on the previously obtained starting point. If the effective value obtained for this range exceeds a predetermined threshold value, it can be considered that there is a kick phenomenon. In step n8 shown in FIG. 2, the determination is performed, and if there is a kick phenomenon, the occurrence of the ground fault is displayed and recorded in step n9 and transmitted to another monitoring device. If no kick phenomenon occurs in any of the phase voltage signals, the zero phase voltage signal detected in step n2 and the starting point obtained in n4 is not due to an instantaneous ground fault, and the following new sampling is performed. The same process is performed again for the data string including the data (n8 → n1 ...).

As described above, when a voltage exceeding the predetermined effective value is generated in the zero-phase voltage signal, and the kick phenomenon occurs in each phase voltage in agreement with the starting point, the instantaneous ground fault occurs. It is detected that it has occurred.

[0027]

According to the present invention, the determination is made not only by the effective value level of the zero-phase voltage signal, but by the AND condition with the presence or absence of the kick phenomenon of each phase voltage. Erroneous evaluation due to the so-called ghost voltage that is superimposed is eliminated, and a ground fault can be detected with high accuracy. Further, according to the present invention, since the rising phenomenon of the ground fault is caught, not only the instantaneous ground fault but also the complete ground fault can be detected.

[Brief description of drawings]

FIG. 1 is a block diagram of a ground fault detection device according to an embodiment of the present invention, and FIG. 2 is a flowchart showing the processing procedure thereof.
FIG. 3 is a waveform diagram of each signal before and after the occurrence of the ground fault. FIG. 4 is a waveform diagram after the zero-phase voltage signal and its differential filter calculation.
FIG. 5 is an explanatory diagram for detecting the starting point. FIG. 6 is a waveform diagram after the phase voltage signal and its digital high-pass filter calculation and difference filter calculation.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Masatsugu Hirose 2109, Yashima Nishimachi, Takamatsu City, Kagawa 8 Shikoku Research Institute Ltd. (72) Takanori Tsunoda 47, Umezu Takaunecho, Ukyo-ku, Kyoto Nissin Electric Co., Ltd. (72) Inventor Hideto Oki 47 Umezu Takaunecho, Ukyo-ku, Kyoto City Nissin Electric Co., Ltd.

Claims (2)

[Claims] 1. A zero-phase voltage determination process of determining an effective value of a zero-phase voltage and determining whether the value exceeds a given threshold value, and an effective value of a high frequency component of each phase voltage, Including a kick phenomenon determination process for determining whether or not the effective value exceeds a given threshold, and detecting a state that satisfies both the conditions of the zero-phase voltage determination and the kick phenomenon determination as a ground fault occurrence state. Ground fault detection method. 2. A difference value of the zero-phase voltage with respect to the value of the fundamental wave frequency before an integral period is calculated, a starting point of the ground fault is calculated from a change in the difference value with the passage of time, and the starting point is used as a reference. The ground fault detection method according to claim 1, wherein a time range for determining the presence or absence of the kick phenomenon is limited.