JP2001183408A - Level measuring method coping with a plurality of measurement frequencies - Google Patents

Abstract

PROBLEM TO BE SOLVED: To put a leakage current detector to common use, without requiring a setting, even in two areas where the frequency of a leakage current is 50 Hz and 60 Hz. SOLUTION: The device shown in Fig. (a) is constituted of a CT(current transformer) 13 for detecting leakage current from a low-voltage cable way, an I/V converter 14 for performing current/voltage conversion, a first AMP (amplifier) 15 for amplifying received voltage, an LPF(low-pass filter) 16 for removing high-frequency components from the received voltage, second and third AMP 17a, 17b for branching the received voltage into two lines, a one-chip CPU(central processing unit) 18 for measuring the level of the received voltage, A/D(angle-to-digital) converters 19a, 19b incorporated in the CPU 18, and a display 20 for displaying the value of the leakage current. The device shown in Fig. (b) is made up of a first LPF 21 for finding the average of an input voltage by sampling it at a prescribed frequency, an oscillator 22 and a frequency divider 23 for determining the sampling frequency for the first LPF 21, and a second LPF 24 for computing the moving averages of input data.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a level measuring method corresponding to a plurality of measuring frequencies, and more particularly to a means for detecting a current value of a commercial power supply leaking to the ground due to insulation deterioration of a circuit on a low-voltage circuit of a power receiving facility. And a level measurement method corresponding to a plurality of measurement frequencies.

[0002]

2. Description of the Related Art With the development of LSIs, the field of application of digital signal processing has expanded, and high-level applications using low-cost microcomputers (hereinafter referred to as μCPU) or digital signal processors (hereinafter referred to as DSP) have been developed. Functional processing has become available. In addition, by realizing functions that have conventionally been realized using a large number of analog parts by digital signal processing, many functions are supported by software, and the cost of the apparatus is reduced. Therefore, a description will be given of a means for detecting a current value at which a commercial power supply leaks to the ground due to insulation deterioration of a power supply path on a low-voltage path of a power receiving facility, using a level measurement method corresponding to a plurality of measurement frequencies.

FIG. 4 is an example of a configuration diagram for detecting a leakage current in a low piezoelectric path. 1 is a power receiving transformer, 2 is a single-phase three-wire low-voltage circuit, 3a and 3b are load devices, 4 is an insulation resistance, and 5 is a current transformer (hereinafter referred to as CT).
And 6 denotes a leakage current detector. The operation of FIG. 4 will be described. The high-voltage path is lowered by the power receiving transformer 1 to the single-phase three-wire low-voltage path 2, and the load devices 3a and 3b are connected. At this time, if the insulation of the electric circuit deteriorates and the insulation resistance 4 decreases, a leakage current of the commercial power supply occurs through the neutral point of the single-phase three-wire low-voltage path 2 grounded at one point. Therefore, the leakage current is extracted by the current transformer 5 provided on the ground line, and the value of the leakage current is detected by the leakage current detector 6.

FIG. 5 is an example of a basic functional block diagram constituting a conventional leakage current detector. The figure shows a low-pass filter (hereinafter referred to as the “low-pass filter”) that extracts
LPF) 7 and 8, a clock oscillator 9, a frequency divider 10 for determining the number of samples in the filtering process, a changeover switch 11 for switching the sampling frequency, and a display 12 for displaying the detected DC voltage value. Constitute. Referring to FIG. 5, when the commercial power frequency is 50
The operation at the time of measuring the level of the leakage current will be described by taking the case of Hz or 60 Hz as an example. First, the LPF is realized using a μCPU, and the LPF has a two-stage configuration to obtain desired performance. Thus, the leakage current of 50 Hz extracted from the low-voltage path is input to the LPF 7 to perform a filtering process. In the LPF 7, the sampling frequency is set to 32 times the input frequency, and 50 Hz × 32 times = 1.
The input data is sampled at 600 Hz. If the frequency of the leakage current is 60 Hz, sampling of input data is performed at a sampling frequency of 60 Hz × 32 times = 1920 Hz. Therefore, a changeover switch 11 is provided to cope with the difference in the frequency of the leakage current,
If it is 0 Hz, the sampling frequency is 1600H
The frequency divider 10 is set so as to be z, and the leakage current is 60 Hz.
, The frequency divider 10 is set so that the sampling frequency becomes 1920 Hz. Next, the sampling is performed 32 times in one cycle over eight cycles, and a filtering operation is performed on 32 × 8 = 256 sampling data. Next, in LPF8, LPF
32 calculation results output from F7 are collected, and a filtering operation by the moving average method is performed on the collected results, and a calculation result having a DC value is output. The detection output having the DC value indicates the value of the leakage current on the display 12.

[0005]

However, the conventional level measuring method corresponding to a plurality of frequencies has the following problems. That is, when this level measurement method is adopted for a leak current detector or the like, the commercial frequency of the leak current, which is a measurement signal, is 50 Hz and 60 Hz in East Japan and West Japan, respectively.
Is different. Therefore, in the A / D converter of the digital filter, which is one of the means for realizing the present level measurement method, if the number of samplings is the same at 50 Hz or 60 Hz, the sampling frequency is the same as when the input signal is 50 Hz. It was necessary to change between 60 Hz. If the number of samples is 32 and the input signal is 50 Hz, the sampling frequency is 1600 H
When z is 60 Hz, the sampling frequency is 1920
Hz. Therefore, in the conventional level measurement method corresponding to a plurality of frequencies, a changeover switch is provided and 50
The sampling frequency is set by a changeover switch when using in the Hz range and when using in the 60 Hz range, so that handling is complicated and the cost of the leakage current detector is increased. The present invention has been made in order to solve the problem of the conventional level measurement method corresponding to a plurality of measurement frequencies as described above, and a leakage current detector in which a frequency of a leakage current is in a 50 Hz band or a 60 Hz band. Even if there is, it is an object of the present invention to provide a level measurement method corresponding to a plurality of measurement frequencies that can be used in both areas without any setting.

[0006]

To achieve the above object, a level measuring method according to the present invention corresponding to a plurality of measuring frequencies has the following configuration. A level measuring method corresponding to a plurality of measuring frequencies according to claim 1, wherein after amplifying a signal under measurement to a desired level value and removing high frequency components,
In a method of performing level measurement using a digital filter functioning as a low-pass filter having a stage configuration, as a filtering process of a first stage of the digital filter, one cycle of a signal under measurement is sampled N times, and the sampled value is then M cycles. The calculated value of the DC component is obtained by adding and calculating an average value. Further, as the filtering operation of the second stage of the digital filter, L filtering operation values of the first stage are combined to calculate a moving average. When the calculated value of the DC component is obtained and output to the display, when the signal under measurement has two unique frequencies, a common sampling frequency is set as an intermediate value between the respective sampling frequencies corresponding to the two frequencies. , And the two frequency signals to be measured are sampled. The level measuring method corresponding to a plurality of measuring frequencies according to claim 2 uses an 8-bit μCPU as a means for realizing the digital filter, and uses two analog / digital converters built in the 8-bit μCPU. The resolution of the filtering operation is configured to be 16 bits.

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail based on illustrated embodiments. 1A and 1B show an embodiment of a level measuring method corresponding to a plurality of measuring frequencies according to the present invention. FIG. 1A shows an example of a configuration of a leakage current detector, and FIG. 1B shows a digital signal used in the present embodiment. Each functional diagram of the filter is shown. As shown in FIG. 1A, a leakage current detector according to the present invention includes a CT 13 for detecting a leakage current of a low piezoelectric path, a current /
An I / V converter 14 for performing voltage conversion, a first AMP 15 for amplifying a reception voltage, an LPF 16 for removing a high-frequency component from the reception voltage, a second AMP 17a for inputting a reception voltage branched into two systems, and a second AMP 17a. Three AMPs 17b, a one-chip CPU 18 for measuring the level of the reception voltage, and C
A / D converters 19a and 19b built in the PU 18,
The display device 20 displays the leakage current value. The operation when the leakage current detector thus configured is connected to the ground line of the electric circuit as shown in FIG. 4 will be described. First, when a leakage current flows to the ground due to insulation failure or the like in the low piezoelectric path, the leakage current is extracted by the CT 13 inserted into the ground line, and is converted into a voltage value by the I / V converter 14. Next, the first amplifier 15 amplifies the received voltage to a level suitable for performing level measurement, and amplifies the LPF1.
6 to attenuate the high-frequency component of the received voltage waveform and remove the adverse effect of the high-frequency component on the filtering process of the digital filter. Further, the output of the LPF 16 is 2
The second amplifier 17a and the third amplifier 1
7b, the outputs of the amplifiers 16a and 16b are
8 and output to the A / D converters 19a and 19b built in.

Here, the reason why the received voltage is branched into two systems and input to the two-channel A / D converter is that (1) the operation in the filtering process uses addition / subtraction and division, and the addition / subtraction is mainly performed and the operation itself is performed. Is not complicated. (2) In order to reduce the hardware cost of digital filters, A / D
This is because a μCPU having two or more converters is used. Further, since the resolution of the μCPU for performing the filtering operation is 8 bits, as means for securing the required 16-bit resolution in the embodiment according to the present invention,
By dividing the amplitude range of the reception voltage into two and allocating two A / D converters to each range, a resolution of 16 bits is obtained. Next, the CPU 18 detects the level of the received voltage by the two-stage LPF realized by the digital filter, and inputs the obtained DC voltage value to the display 20 for display.

FIG. 1B is a block diagram of the CPU 18 included in the CPU 18.
6 is an embodiment showing a configuration of a digital filter realizing an LPF having a stage configuration. The digital filter samples the input voltage at a predetermined frequency to obtain an average value, a first LPF 21 for determining a sampling frequency of the first LPF 21, an oscillator 22 and a frequency divider 23, and calculates a moving average value of input data. And the second LPF 24 to be used. First, the reason why two stages of LPFs are provided will be described.
The memory required for the calculation by the μCPU is used to constitute the digital filter using only the built-in memory. In this embodiment, the memory used for the calculation is:
There are a total of 66 bytes of 2 bytes in the first stage and 2 bytes in the second stage × 32 = 64 bytes. If this digital filter is composed of only one stage of the moving average method, 2 bytes × 256 × 32 = 16384 bytes are required as a memory for calculation. By the way, R of 1-chip μCPU
The capacity of the AM is generally about 256 to 2 kbytes, of which about 20% of the RAM capacity can be used for filtering operation and the like.

Next, an outline of the operation of the digital filter according to this embodiment will be described. In the present embodiment, the frequency of the measurement signal is two frequencies of 50 Hz and 60 Hz. However, if the measurement signal of two frequencies can be sampled at one common sampling frequency and converted into a digital signal, Even if the frequency is 50 Hz or 60 Hz, there is no need to set the measurement circuit, and the measurement circuit can be shared by both frequencies. Therefore, usually, the frequency obtained by dividing the sampling frequency by the number of samples is, for example, 1600 HzＨ32 = 50H if the frequency of the measurement signal is 50 Hz.
192 if the frequency of the measurement signal is 60 Hz.
Although there is a relationship of 0 Hz 、 32 = 60 Hz, a sampling frequency common to both is 176 which is an intermediate value between the two.
If the frequency is 0 Hz, the frequency obtained by dividing the sampling frequency by the number of samples is 1760 Hz ÷ 32 = 55 Hz. When the frequency of the measurement signal is 50 Hz, an error of + 10% is given.
In the case of 0 Hz, an error of -10% is provided. As described above, as a phenomenon when the sampling frequency has an error, the output signal of the digital filter fluctuates, and different output characteristic errors occur depending on the magnitude of the error of the sampling frequency. This error characteristic is considered to occur in the course of the filtering process in the CPU, and the A / D
This is presumed to be due to the fact that the sampling frequency is limited to a certain value or less at the time of conversion, and that the arithmetic processing is performed using a finite word length. By the way,
When the simulation is performed by changing the sampling number for the input data of the first-stage LPF from 32 to 64, the error of the level characteristic output by the first-stage LPF is 1 /
2, but requires a large amount of memory and is not an option.

FIG. 2 shows that the measurement signal has a frequency of 50H.
The first LPF when the sampling frequency of the first LPF has an error when the leakage current is 10 mA in z.
And error characteristics output by the second LPF. The figure shows a value obtained by dividing the sampling frequency by the number of samples, which is a value that is originally 50 Hz. The figure shows a case where the error is changed from + 0.0002% to + 10%. Is the output characteristic of the first LPF,
The period during which the output characteristics of the first LPF fluctuates, the output characteristics of the second LPF, and the width over which the output characteristics of the second LPF fluctuate, respectively. It can be seen from the result that if the sampling frequency has an error, the first L
The output characteristic of the PF causes an error, and the error has its own fluctuation. Further, the fluctuation period of the fluctuation becomes shorter as the error given to the sampling frequency becomes larger. Therefore, the error of the sampling frequency is + 2% or + 10%
The beat cycle of the output characteristic shown in the case where the first LPF is provided is shortened, and the fluctuation of the output characteristic of the first LPF is reduced by the second LPF.
Attenuated sufficiently at F, and no fluctuation has occurred from the output characteristics of the second LPF. On the other hand, when the error of the sampling frequency is given by + 0.0002% or + 0.002%, the beat period of the output characteristic becomes longer, and the fluctuation of the output characteristic of the first LPF is attenuated by the second LPF. However, the output characteristic of the second LPF also fluctuates. The fixed output error of the second LPF does not pose a problem because it can be corrected at the time of the filtering operation. Therefore, as described above, if an error is caused in the sampling frequency, an error occurs in the output characteristics of the digital filter. However, if the error in the sampling frequency is reduced to about 10%, the error in the output characteristics of the next-stage filter is sufficient. The sampling frequency can be eliminated, and an error is given to the sampling frequency of + 10% for the measurement signal of 50 Hz as in the present embodiment, and an error of −10% for the measurement signal of 60 Hz. A / D conversion can be performed using the sampling frequency.

The operation of FIG. 1B will be described with respect to a case where the measurement signal is 50 Hz. In the first LPF 21, a frequency corresponding to an input frequency obtained by dividing a sampling frequency for input data by the number of samples and an actual input frequency are obtained. Set to have an error with the frequency. In this embodiment, the frequency obtained by dividing the sampling frequency by the number of samples with respect to the input frequency of 50 Hz is 55 Hz, and +1
Assuming that the number of samples is 32 in order to have an error of 0%, the sampling frequency is 1760 Hz. Therefore, the first LPF 21 samples the input voltage value 32 times in one cycle, and further adds eight cycles of sampling data. Therefore, based on a total of 256 sampling data, 1 計 55 Hz × 8 ≒ 181 msec Each time a filtering operation is performed, an average value is obtained.

Next, in the second LPF 24, a moving average is obtained as a filtering operation.
The data output from the PF 21 is input every 181 msec, and an average value is calculated from 32 pieces of data. The calculation method is a moving average method, in which every time new data is input, the oldest data is discarded to shift the data to be calculated in order, and the average value is calculated each time. Therefore, the second LPF 24 is 181 mse
Output data for each c.

FIG. 3 is an example of a flow chart showing the processing steps of the digital filter according to the present invention. The numerical values described are values when the sampling frequency is shifted by + 10%. Referring to the figure, in the A / D conversion process, 50 Hz input data is sampled 32 times in one cycle, and the data is output about every 707 μsec. Also, 8
In the A / D conversion processing of the bit configuration, the amplitude range of the reception voltage is divided into two, and two A / D converters are assigned to each range, thereby obtaining a resolution of 16 bits. Next, in the process of adding 256 times, about 707 μm is obtained by A / D conversion processing.
256 data output every second are added, and 0.7
The average value is calculated and output every 07 msec × 256 × 181 msec. The buffer for addition is prepared by 2 bytes and 16 bytes.
Performs a bit operation. Therefore, every 181 msec, 16
The calculation result data output in bits is sequentially input to a moving average buffer prepared by 2 bytes × 32 = 64 bytes in order to perform a moving average process of 32 pieces. In the moving average process, the moving average process is performed every time data is input every 181 msec, the oldest data is discarded each time new data is input, and a 16-bit moving average value is calculated each time. The calculation result is output as 16-bit data every 181 msec, and digital display processing of three digits is performed. Although the above description has been made for the case where the measurement frequency is 50 Hz, the same operation is performed even when the measurement frequency is 60 Hz, and the leakage current to be measured is 50 Hz or 60 Hz.
Operates at the same sampling frequency regardless of

[0015]

As described above, according to the present invention, when the leakage current detector is used, even if the commercial power supply frequency is in the 50 Hz range or the 60 Hz range, the same operation can be performed without any change of the setting. Can be shared, so that a great effect can be exhibited in the operation of the leakage current detector.

[Brief description of the drawings]

FIG. 1 shows an embodiment of a level measurement method corresponding to a plurality of measurement frequencies according to the present invention.

FIG. 2 is a table showing error characteristics obtained as a result of simulating the output characteristics of a digital filter according to the present invention using a computer.

FIG. 3 is an example of a flowchart showing a process of processing a digital filter according to the present invention.

FIG. 4 is an example of a configuration diagram for detecting a leakage current of a low piezoelectric path.

FIG. 5 is an example of a basic functional block diagram of a conventional leakage current detector.

[Explanation of symbols]

 1. Power receiving transformer, 2. Single-phase three-wire low-voltage circuit, 3a, 3b. Load equipment, 4. Insulation resistance, 5. CT, 6. Leakage current detector, 7, 8. LPF , 9-clock oscillator, 10-divider, 11-changeover switch, 12-display, 13-CT, 14-I / V converter, 15-AMP, 16-LPF, 17a, 17b AMP, 18 CPU, 19a, 19b A / D converter, 20 display, 21 LPF, 22 oscillator, 23 divider, 24 LPF

Claims (2)

[Claims] 1. A method for amplifying a signal to be measured to a desired level value and removing a high-frequency component, and then performing a level measurement using a digital filter functioning as a two-stage low-pass filter. In the first stage of filtering processing, one cycle of the signal to be measured is sampled N times, the sampled value is added for M cycles to calculate an average value, thereby obtaining an operation value of a DC component. As the filtering operation of the first stage, when the calculated value of the DC component is obtained by summing L filtering operation values of the first stage and calculating a moving average and outputting the calculated value to the display, the signal under measurement has two unique frequencies. At one time, a common sampling frequency is set as an intermediate value between the respective sampling frequencies corresponding to the two frequencies, Level measurement method corresponding to a plurality of measurement frequencies, characterized in that samples the signal to be measured in frequency. 2. An 8-bit microcomputer is used as a means for realizing the digital filter, and the resolution of the filtering operation is set to 16 bits by using two analog / digital converters built in the 8-bit microcomputer. 2. The level measurement method according to claim 1, wherein the level measurement is performed for a plurality of measurement frequencies. .