KR100645113B1 - Noise reduction and Quantity decision method in partial discharge measurement - Google Patents

Abstract

(여기서, Qunatity는 부분방전 측정량, N은 부분방전측정데이타의 반복측정회수, M은 60Hz 전원의 1주기동안 측정된 부분방전 펄스 데이터 수, Q는 부분방전 펄스레벨을 정수화시켰을때의 최대값, p, φ는 각각 부분방전 펄스의 크기 및 위상, w_n,w_p는 부분방전신호의 증폭이득에 따라 결정된 가중치 상수이다.)에 의하여, 정량화시킨다.The present invention relates to a method for removing noise and quantifying a signal accurately and reproducibly, which effectively removes noise included in a partial discharge measurement signal and enables automatic trend management based on measurement data. In the method for removing noise and quantifying the partial discharge measurement signal according to the present invention, the partial discharge signal is repeatedly measured at a predetermined period to collect the partial discharge measurement data, and then processed into phase-size-number data, and the processed phase After detecting the low frequency data from the size-number data as noise, removing the detected noise from the partial discharge measurement data,

Where Qunatity is the partial discharge measurand, N is the number of repetitive measurements of the partial discharge measurement data, M is the number of partial discharge pulse data measured during one cycle of 60 Hz power supply, and Q is the maximum value when the partial discharge pulse level is integerized. , p and φ are the magnitude and phase of the partial discharge pulse, respectively, and w _n , w _p are weight constants determined according to the amplification gain of the partial discharge signal.

Partial discharge, noise, quantification, frequency, trend management,

Description

Noise reduction and Quantity decision method in partial discharge measurement

1 is a block diagram showing the configuration of a partial discharge measurement system.

2 is a control flowchart of a method for removing noise and quantifying a partial discharge measurement signal according to the present invention.

3A is a table showing partial discharge measurement data, and FIG. 3B is a phase-time-size graph derived from the partial discharge measurement data.

4A is a table showing phase-size-count data, and FIG. 4B is a phase-size-count graph derived from the table.

5 is a phase-size-number graph in which noise is removed by the present invention.

6 is a phase-time-size graph from which noise is removed by the present invention.
* Description of the symbols for the main parts of the drawings *
10: cable connection 11, 12: antenna
13, 14: first and second partial discharge antennas 15, 16: first and second preprocessing amplifiers 17, 18: first and second band pass filters
19, 20: first and second integral and maximum value acquisition unit 22: analog switch
23: analog to digital conversion unit 24: digital signal processing unit
25: data sending section 26: sound reading section
27: display alarm unit

The present invention is a partial discharge measurement system used for the diagnosis of the health of electrical equipment, accurate and reproducibility, which effectively removes noise included in the partial discharge measurement signal, and enables automatic trend management based on measurement data The present invention relates to a method for removing noise and quantifying an excellent partial discharge measurement signal.

In general, electrical equipment such as power cables, transformers, transformers for measuring voltage or current, breakers, and the like are aging, and the insulation is weakened, causing partial discharge. Therefore, partial discharge signals are often used for the diagnosis of the health of electrical installations.

As a conventional apparatus for diagnosing the integrity of an electrical installation using such partial discharge, an example of a partial discharge measuring system is shown in FIG.

1 is a partial discharge measuring system disclosed in published patent application 2004-0020356 (published on March 9, 2004) filed by the inventor of the present invention. Referring to the drawings, the partial discharge measuring system is The first and second partial discharge measuring antennas 13 and 14 and the first and second partial discharge measuring antennas 13 and 14 which are positioned at both edges of the cable connecting part 10 are respectively connected to the first and second partial discharge measuring antennas 13 and 14. A first and second band pass filters 17 and 18 to which second preprocessing amplifiers 15 and 16, the first and second preprocessing amplifiers 15 and 16 are connected, respectively, and the first and second bands. First and second integral and maximum value acquisition units 19 and 20 to which pass filters 17 and 18 are connected, and analog switches to which the first and second integration and maximum value acquisition units 19 and 20 are connected. (22), the analog-digital converter 23 connected to the analog switch 22, and the analog-to-digital converter 23 A digital signal processor 24, a data transmitter 25 connected to the digital signal processor 24, a health reader 26 connected to the data transmitter 25, and the health reader 26 It is composed of a display alarm unit 27 connected to.

The configured partial discharge measurement system detects partial discharge radio waves leaking from the cable connection part 10 through the first and second partial discharge measurement antennas 13 and 14, and the first and second preprocessing amplifiers 15 and 16. ) And first and second band pass filters 17 and 18 are amplified and filtered, and the first and second integrating and maximum value obtaining units 19 and 20 obtain the maximum value from the integral waveform of the partial discharge measurement signal. Detect. This is a process for measuring the partial discharge amount by using the integrated value rapidly increasing when the electromagnetic signal is generated and gradually decreasing the integrated value when the electromagnetic wave is reduced.

Accordingly, the obtained integral maximum value is converted into a digital value by the analog-to-digital converter 23 and applied to the digital signal processor 24. The digital signal processor 24 removes abnormal signals and transmits normal measurement data. The unit 25 is sent to the health reading unit 26 through the unit 25. The health readout unit 26 performs a fast fourier transform (FFT) of the measured data input as described above, and analyzes the frequency component to perform storage and trend analysis.

At this time, if the measured value is greater than the maximum allowable value (for example, a value corresponding to 500 pC partial discharge) or shows a tendency to continuously increase, it is determined to be abnormal and informs the operator through the display alarm unit 27.

However, as described above, the partial discharge signal used to determine the soundness of the electrical equipment is very small (a few mV to several hundred mV), and thus is affected by ambient noise and shows an irregular occurrence frequency. Removal and accurate trend management are difficult.

In the related art, by using the partial discharge measurement system as shown in FIG. 1, the partial discharge is repeatedly measured and processed only until a phase-size-recovery graph is generated, and then the operator observes the shape of the generated graph. The trends were analyzed by subjective experience. Therefore, accurate trend analysis was difficult for the skilled person.

In addition, as in the partial discharge measurement system shown in FIG. 1, when a certain abnormal value is measured, an alarm is generated, but a malfunction often occurs due to the influence of ambient noise, and an average value of the measured value is used for the alarm. There is a problem that the reproducibility and reliability of the measurement data are inferior.

The present invention has been proposed in order to solve the above-mentioned conventional problems, and an object thereof is to accurately remove noise included in a partial discharge measurement signal and to enable automatic trend management based on measurement data. The present invention provides an excellent method for removing noise and quantifying an excellent partial discharge measurement signal.

(여기서, Qunatity는 부분방전 측정량, N은 부분방전측정데이타의 반복측정회수, M은 60Hz 전원의 1주기동안 측정된 부분방전 펄스 데이터 수, Q는 부분방전 펄스레벨을 정수화시켰을때의 최대값, p, φ는 각각 부분방전 펄스의 크기 및 위상, w_n,w_p는 부분방전신호의 증폭이득에 따라 결정된 가중치 상수이다.)에 의하여, 상기 잡음이 제거된 위상-크기-횟수 데이타를 부분방전 측정량으로 정량화하는 정량화 단계를 포함하여 이루어진다.As a constituent means for achieving the object of the present invention described above, the method for removing noise and quantification of the partial discharge measurement signal according to the present invention, the collection step of collecting the partial discharge measurement data by repeatedly measuring the partial discharge signal at a predetermined period ; A processing step of processing the collected measurement data into phase-size-number data when a predetermined number of partial discharge measurement data are collected; A noise removing step of detecting data having a low occurrence frequency from the processed phase-size-number data as noise and removing the detected noise from partial discharge measurement data; And quantifying the partial discharge measurement data from which the noise is removed.

Where Qunatity is the partial discharge measurand, N is the number of repetitive measurements of the partial discharge measurement data, M is the number of partial discharge pulse data measured during one cycle of 60 Hz power supply, and Q is the maximum value when the partial discharge pulse level is integerized. , p and φ respectively represent the magnitude and phase of the partial discharge pulse, and w _n and w _p are weight constants determined according to amplification gain of the partial discharge signal. And a quantification step of quantifying the discharge measurand.

In the noise elimination and quantification determination method of the partial discharge measurement signal according to the present invention, the collecting step is characterized in that the step of storing the partial discharge measurement data as the pulse size for each phase of the partial discharge measured in each period.

Further, in the method for removing noise and quantifying the partial discharge measurement signal, the phase-magnitude-count data of the processing step may be the number of partial discharge pulses having the same phase and the same magnitude.

Further, in the method for removing noise and quantifying the partial discharge measurement signal, the noise removing step may include a phase and a partial discharge pulse having a frequency less than the maximum number of occurrences of irregularly distributed partial discharge pulses. It is characterized in that the step of separating from the size-count data, and removing it from the partial discharge measurement data.

In addition, the quantification step is set such that w _p is lower as the gain of the partial discharge signal increases, w _n is less than or equal to 1.0, and is set to have the same w _p and w _n for the same amplification gain. do.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description and the accompanying drawings, the same reference numerals are used for components having the same functions.

2 is a flowchart showing an embodiment of a method for removing noise and quantifying a partial discharge measurement signal according to the present invention.

The noise canceling and quantification determination method of the partial discharge measurement signal according to the present invention is configured as part of a data processing program in the partial discharge measurement system as shown in FIG. 1, for example, the health readout unit 26 and It can be implemented in the form of a program in a personal computer (PC) equipped with a window operating system such as the display alarm unit 28. In addition, in the following, the partial discharge measurement data is sensed by the antennas 11 and 12 in the partial discharge measurement system as shown in FIG. 1 and transmitted by the data transmitter 25 after amplification, filtering, and digital conversion. As the measured data, the present invention relates to the processing of the partial discharge measurement data measured by the method of Fig. 1 and other methods.

The method for removing noise and quantifying a partial discharge measurement signal according to the present invention begins with collecting and storing partial discharge measurement data at predetermined periods.

Referring to FIG. 2, the partial discharge measurement system as shown in FIG. 1 and the state in which the data measured by the system are initialized 201 and 202 are detected, and the measurement system detects the partial discharge signal at a predetermined period to generate a predetermined type of data. It may be implemented by collecting and storing the converted to (203). At this time, the collected data is collected by a set number of times, and the partial discharge measurement data collected by the set number of times is expressed as a pulse size value for each phase of the partial discharge measured at each period. That is, the collected partial discharge measurement data is stored as a table displaying the magnitudes v (N, M) of the partial discharge pulses for each phase and the number of measurement, as shown in FIG. 3A. When the partial discharge measurement data shown in FIG. 3A is represented by a phase-time-size graph, it appears as shown in FIG. 3B.

When the partial discharge measurement data as described above is collected N times (N is randomly set to one or more natural numbers) (204), the partial discharge measurement data collected during the N times is processed into phase-size-number data (205). ). The phase-magnitude-count data is the number of partial discharge pulses having the same phase and the same magnitude. In the processing step 205, the number of partial discharge pulses of the same phase and magnitude is calculated in the table as shown in FIG. Counted and represented by phase and magnitude tables. FIG. 4A is a table showing the results of processing the phase-magnitude-number data in step 205, and is expressed as the number of partial discharge pulses n (Q, M) for each phase (phi) and magnitude (p). If the table shown in FIG. 4A is expressed as a phase-size-recovery graph, the table shown in FIG.

As shown in FIG. 4B, when processing the phase-size-number data, the partial discharge pulses with low incidence are distinguished. In addition, a pulse generated at a frequency lower than the occurrence frequency of the partial discharge pulse is likely to be noise. Accordingly, the pulse with the low number of occurrences is identified in the processed phase-size number data (206).

In operation 206, the low frequency pulse value identified in step 206 is removed from the collected partial discharge measurement data.

In general, the maximum number of occurrences of the partial discharge pulses having an irregular distribution is twice when measured at 300 times or less, and the measurement data having a lower frequency is removed as noise. 5 is a graph showing a state in which the low frequency of noise is removed from the phase-size-recovery data shown in FIG. 4B, and FIG. 6 is a state in which the low frequency of noise is removed from the partial discharge measurement data shown in FIG. 3B. This graph shows

According to the above, noise included in the partial discharge measurement data can be effectively removed, thereby preventing malfunction caused by the noise.

In addition, the present invention substitutes the noise-reduced partial discharge measurement data into the equation shown in Equation 1 below to quantify the partial discharge amount (208).

In Equation 1, Qunatity is the partial discharge measurement amount, N is the number of repeated measurements of the partial discharge measurement data, M is the number of partial discharge pulse data measured during one cycle of a 60 Hz power source, Q is when the partial discharge pulse level is integerized The maximum value of, p, φ is the magnitude and phase of the partial discharge pulse, respectively, w _n , w _p is a weight constant determined according to the gain of the partial discharge signal.

In the above, w _p is set to be lower as the amplification gain of the partial discharge signal becomes larger, and w _n is set to a value of 1.0 or less, and is set to have the same w _p and w _n for the same amplification gain.

When the partial discharge measurement data is quantified as the partial discharge amount as described above, the quantification result can be more accurate and reliable by the average value as in the prior art, and trend management can be performed based on such quantification.

Hereinafter, an embodiment of a method for removing noise and quantifying a partial discharge measurement signal according to the present invention will be described.

First, when the partial discharge measurement data is measured by dividing one cycle (16,667 msec) of the 60 Hz power supply by M, M data are obtained during one measurement, which is measured N times. 3B is a method of obtaining 128 data during the first cycle (16,667 msec), and is a partial discharge measurement data collected by measuring 288 data.

The partial discharge measurement data of FIG. 3B is converted into phase-size-number data as shown in FIG. 4B. The partial discharge pulse having the lowest frequency of occurrence is removed from the phase-size-number data of FIG. 4B, and the result is shown in the graph of FIG. 5. Referring to the graph of FIG. 5, it can be confirmed that a low frequency pulse has been removed.

Next, FIG. 6 converts the phase-magnitude-number data of FIG. 5 into phase-time-magnitude data, and it can be seen that pulse data showing an irregular and low occurrence frequency is removed from the graph of FIG. 6.

When the partial discharge measurement data is quantified by substituting Equation 1, M = 128, N = 288, and Q = 255. Here, Q = 255 is a case where the magnitude of the partial discharge pulse data is represented by 8 bits, and in this case, the measured value has 0 to 255.

On the other hand, the present invention is not limited to the above description, various modifications and applications can be made within the technical scope of the present invention.

As described above, according to the method for removing and quantifying the noise of the partial discharge measurement signal of the present invention, it is possible to effectively remove the ambient noise affecting the partial discharge measurement signal, thereby increasing the reliability of the measurement result, and adding a plurality of measurements. By quantifying more accurate partial discharge amount from the data, it is possible to improve reproducibility and to manage trend of the measurement data.

Claims (5)

Collecting the partial discharge measurement data by repeatedly measuring the partial discharge signal at a predetermined cycle; A processing step of processing the collected measurement data into phase-size-number data when a predetermined number of partial discharge measurement data are collected; A noise removing step of detecting data having a low occurrence frequency from the processed phase-size-number data as noise and removing the corresponding data from the partial discharge measurement data; And Quantification formula of the partial discharge measurement data from which the noise is removed

Where Qunatity is the partial discharge measurand, N is the number of repetitive measurements of the partial discharge measurement data, M is the number of partial discharge pulse data measured during one cycle of 60 Hz power supply, and Q is the maximum value when the partial discharge pulse level is integerized. , p and φ are the magnitude and phase of the partial discharge pulse, respectively, and w _n and w _p are weight constants determined according to the gain of the partial discharge signal.) And quantifying the noise-rejected phase-size-number data by the partial discharge measurement amount. The method of claim 1, wherein said collecting step And storing the partial discharge measurement data as a pulse size for each phase of the partial discharge measured in each period. The method of claim 2, wherein the phase-magnitude-count data of the processing step is A method for removing noise and quantifying a partial discharge measurement signal, characterized in that the number of partial discharge pulses having the same phase and the same magnitude. The method of claim 3, wherein the noise canceling step Partial discharge pulses of phase and magnitude having a number of occurrences lower than the maximum number of irregularly distributed partial discharge pulses are identified in the phase-size-number data, and the partial discharge pulses of the corresponding phase and magnitude are measured in the partial discharge measurement data. Noise canceling and quantification method of the partial discharge measurement signal, characterized in that for removing the step. According to claim 1, In the quantification formula of the quantification step, w _p is lowered as the amplification gain of the partial discharge signal increases, and w _n is set to be less than or equal to 1.0, and noise cancellation of the partial discharge measurement signal characterized by having the same w _p and w _n for the same amplification gain. And quantification determination method.

Images