CN115166450A - Partial discharge localization method for switchgear components based on pulse current signal spectrum - Google Patents

Abstract

The invention discloses a partial discharge positioning method for switch cabinet components based on pulse current signal spectrum, which belongs to the technical field of partial discharge positioning. The method includes step 1: conducting partial discharge test and measurement of the switchgear, establishing a database of the average test frequency f _ath and the test equivalent frequency width F _wth of the pulse current signal, and establishing a database of the median value of the spectral characteristic; step 2: on-site local Discharge signal detection, collect the partial discharge pulse current signal in the switch cabinet, calculate the measured average frequency f _ats and the measured equivalent frequency width F _wts ; Step 3: Calculate the partial discharge pulse current signal collected in step 2 and the median value of the data in step 1 The Euclidean distance D _i between them is used to determine the partial discharge components according to the principle of minimum distance. The method of the invention can greatly improve the localization accuracy of the on-site partial discharge, prevent the equipment from being put into operation with defects, and greatly promote the progress of the electrical equipment test and detection technology.

Description

Partial discharge localization method for switchgear components based on pulse current signal spectrum

technical field

The invention relates to the technical field of partial discharge localization, in particular to a partial discharge localization method for switchgear components based on a pulse current signal spectrum.

Background technique

As a very important electrical equipment in the power distribution network of the power system, the high-voltage switchgear is mainly used to switch, control and protect the electrical equipment in the process of power generation, transmission, distribution and power conversion in the power system. The components in the switch cabinet are mainly composed of circuit breakers, isolating switches, load switches, operating mechanisms, transformers and various protection devices. During the production, transportation, installation and maintenance of high-voltage switchgear, there are inevitably defects such as cracks, metal burrs and scattered metal impurities. With the long-term operation of the switchgear, these defects cause the local field strength on the insulating surface or inside to be higher than the critical field strength of the insulating medium itself, thus causing the occurrence of partial discharge. At the same time, the long-standing partial discharge phenomenon will further deteriorate the insulation and accelerate the insulation failure. Therefore, strengthening the detection and diagnosis of partial discharge is an important prerequisite to ensure the safe and reliable operation of the switchgear.

Partial discharge detection and diagnosis and positioning technology of high-voltage switchgear is a hot research topic at home and abroad. The generation of partial discharge will be accompanied by various physical phenomena such as sound, electricity, light, heat, etc. Various partial discharge detection methods are developed by monitoring these physical quantities reflecting the discharge phenomenon. The commonly used detection methods at home and abroad are mainly ultrasonic method, ultra-high frequency method and transient ground voltage method. In these methods, partial discharge location is generally carried out by the magnitude of amplitude. These methods are all external detection methods. Due to the complex propagation process of partial discharge signals and complex on-site interference, these methods have low positioning accuracy and large errors. Under good detection conditions, these methods can only determine which side of the switchgear the partial discharge is located, but cannot determine which part of the switchgear the partial discharge is located in. Inaccurate positioning cannot accurately guide maintenance decisions. Therefore, component-level precise localization of PD in switchgear is the focus of future research.

The existing partial discharge positioning methods for switchgear are as follows:

(1) Partial discharge localization method based on the combination of sound and electricity and the combination of sound and sound

The acoustic-acoustic joint time-delay localization method is to arrange four or more ultrasonic sensor nodes in space, and calculate the discharge point position according to the propagation speed of ultrasonic waves in the air and the signal timing measured by each measurement point; the acoustic-electrical joint time-delay localization method One of the ultrasonic sensor nodes in the harmonic-sound joint localization method is replaced by a UHF sensor or TEV sensor node and used as a reference trigger node, and the discharge point position is calculated in the same way. However, this method has poor anti-interference ability, and the positioning is prone to deviation.

(2) Localization method of partial discharge in switchgear based on ground wave principle

During the propagation of the ground wave on the surface of the metal wall, a TEV signal will be induced, and multiple TEV sensors will be used for detection to detect the strength of the TEV signal and the trigger time of the first TEV pulse, which can be realized according to the amplitude and delay. Partial discharge localization in switchgear. However, the ground wave signal attenuates quickly, and the measured signal overlaps with the frequency range of the on-site interference, so the on-site application effect is not good.

(3) Localization of typical partial discharge defects in high-voltage switchgear based on UHF method

Place the two UHF sensors in two different positions. The UHF signal generated by partial discharge propagates to the sensor in the shortest air path. After converting the electromagnetic wave signal into an electrical signal, its waveform information is displayed on the oscilloscope display. superior. The time difference can be obtained from the comparison on the oscilloscope display. According to the time difference and the known wave speed, the straight-line distance difference between the partial discharge source and the two sensors can be calculated, that is, the distance from the discharge source to the vertical bisector of the two sensors. The location of the PD source can be determined by placing sensors at different locations and making multiple measurements. However, at present, the gap of the switch cabinet is small, and the UHF signal is difficult to transmit, so the positioning accuracy is low.

With the increasing attention of the PD detection and localization work group, the related sensing, acquisition and analysis technologies have been vigorously developed. The partial discharge pulse current detection technology of switchgear is an emerging detection method in recent years. In this method, the pulse signal is transmitted through the coupling capacitor connected to the high-voltage bus, and the pulse current signal is received by the sensor connected to the grounding end of the coupling capacitor. The coupling capacitor and partial discharge sensor installed in each cabinet can be used to detect one by one, and the switch cabinet where the partial discharge defect is located can also be judged by the signal amplitude. In addition, based on the influence of switchgear components on the propagation characteristics of the pulse current signal, the component where the partial discharge defect is located can also be determined by analyzing the spectral characteristics of the pulse current signal.

To sum up, the existing switchgear partial discharge positioning technology adopts the time domain positioning method, which is based on the amplitude and time delay of the time domain waveform, but this method requires extremely high resolution of the measuring instrument, and due to interference If there is, it is likely to cause misjudgment and omission of information such as the amplitude of the waveform and the arrival time of the first wave, and the positioning accuracy will be reduced. Therefore, it is necessary to locate and analyze the partial discharge of switchgear components in the frequency domain based on the partial discharge localization method of the switchgear components based on the pulse current signal spectrum.

SUMMARY OF THE INVENTION

The purpose of the present invention is to propose a partial discharge localization method for switchgear components based on pulse current signal spectrum, characterized in that the method comprises the following steps:

Step 1: Carry out the partial discharge test and measurement of the switchgear, establish the database of the test frequency average value f _ath and the test equivalent bandwidth F _wth of the pulse current signal, and calculate the data median value of the partial discharge of each component, and establish the spectral characteristic quantity value database;

Step 2: perform partial discharge signal detection on site, collect partial discharge pulse current signals in the switch cabinet, and calculate the measured average frequency f _ats and the measured equivalent frequency width F _wts ;

Step 3: Calculate the Euclidean distance D _i between the partial discharge pulse current signal collected in step 2 and the median value of the data in step 1, and determine the partial discharge component according to the principle of minimum distance.

The step 1 specifically includes the following sub-steps:

Step 11: Connect the coupling capacitor in parallel in the switch cabinet, and connect the pulse current coupling unit to its ground terminal to couple the pulse current signal; then carry out partial discharge tests on each component in the switch cabinet, and measure the pulse current signal time domain waveform I(t );

Step 12: Perform Fourier transform on the pulse current signal time-domain waveform I(t) in step 11 to obtain the pulse current signal spectrum waveform I(jω);

Step 13: Extract the test frequency mean value f _ath and the test equivalent frequency width F _wth through the pulse current signal spectrum waveform I(jω), and perform multiple tests at the same position, calculate the median value of the data, and establish a median value database of spectral feature quantities .

The step 3 is as follows:

The spectral feature value median database is formed into a set consisting of n vectors M _i ; the vector M _i is:

M _i =(f _athmi ,F _wthmi ), i=1,2,...,n;

Among them, f _athmi is the mean test frequency of the ith component, and F _wthmi is the test equivalent bandwidth of the ith component;

The measured frequency mean f _ats and the measured equivalent bandwidth F _wts form a vector S, namely:

S=(f _ats ,F _wts )

Calculate the Euclidean distance Di between S and _Mi one by one through the Euclidean distance formula, and select the component with the smallest Euclidean distance as the partial discharge component; the calculation method of the Euclidean distance D _i is _:

The beneficial effects of the present invention are:

1. The method of the present invention can greatly improve the accuracy of on-site partial discharge positioning, prevent equipment from being put into operation with defects, and greatly promote the progress of electrical equipment testing and detection technology; and the positioning results can be accurate to a certain component inside the switch cabinet, robust higher sex

2. After accurate positioning is realized based on the method of the present invention, the accuracy and work efficiency of equipment insulation defect diagnosis and prediction, operation and maintenance will be greatly improved, and the safety and reliability of the switch cabinet will be greatly improved;

3. The present invention reduces the probability of insulation failure of high-voltage switch cabinets of 10kV to 35kV, and has significant direct and indirect economic benefits.

Description of drawings

Fig. 1 is the electrical connection diagram in the high-voltage switchgear of the switchgear component partial discharge location method based on the pulse current signal spectrum of the present invention;

FIG. 2 is a flow chart of the partial discharge localization method for switchgear components based on the pulse current signal spectrum of the present invention.

Detailed ways

The present invention proposes a method for locating partial discharge of switchgear components based on the frequency spectrum of the pulse current signal. The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

When a partial discharge occurs in a component in the switch cabinet, when the pulse current signal reaches the detection coupling device, its transfer function depends on the line in the propagation path and the impedance characteristics of each component; the spectral characteristics of the signal obtained by the coupling device are related to the partial discharge. The position is directly related; the part where the partial discharge is located can be mapped by using the spectral characteristics of the partial discharge pulse current signal.

The electrical wiring diagram of the internal components of the high-voltage switchgear is shown in Figure 1. The high-voltage switchgear is mainly composed of high-voltage busbar 1, circuit breaker 2, current transformer 3, arrester 4, live display device 5, grounding switch 6, and high-voltage cable 7. In order to couple the partial discharge pulse signal, a coupling capacitor 8 and a pulse current coupling unit 9 are arranged inside the switch cabinet.

In order to realize accurate localization of partial discharge of switchgear, the present invention proposes a flow chart of partial discharge localization method of switchgear based on the spectral characteristics of pulse current signal, as shown in FIG. 2 . The implementation steps include: first, carry out the partial discharge test and measurement of the switchgear in the laboratory, establish a database of the frequency mean (f _ath ) and equivalent bandwidth (F _wth ) of the pulse current signal, and calculate the partial discharge of each component. The median value of the data; second, carry out partial discharge signal detection on site, collect partial discharge pulse current signals in the switch cabinet, and calculate spectrum parameters, including: frequency mean (f _ats ), equivalent bandwidth (F _wts ); third, Calculate the Euclidean distance between the sampling parameter and the median value of the database standard sample, and determine the part where the partial discharge is located according to the principle of minimum distance. details as follows:

(1) Establish a database of partial discharge pulse current spectrum parameters of high-voltage switchgear components

Partial discharge test of each component of the switchgear and acquisition of pulse current signal time-domain waveform; according to the actual case of partial discharge of the switchgear, make partial discharge defects in each part of the switchgear. A coupling capacitor is connected in parallel in the switch cabinet, and a coupling unit is connected to its ground terminal to couple the pulse current signal. Partial discharge tests are carried out for each component in the switch cabinet, and the time domain waveform I(t) of the pulse current signal is measured.

Fourier transform and spectrum parameter calculation of pulse current signal; Fourier transform is performed on pulse current signal I(t) to obtain the spectrum waveform I(jω) of pulse current signal. The Fourier transform formula is:

The characteristic quantity of the frequency domain waveform I(jω) of the pulse current signal is represented by the mean value and dispersion of the spectrum waveform. These two characteristic quantities are called: the frequency mean value (f _a ) and the equivalent bandwidth (F _w ). Taking |I(jω)| ² as the probability density, the frequency domain characteristic parameters are obtained by calculating its mean and standard deviation. The frequency mean (f _a ) and the equivalent bandwidth (F _w ) are respectively defined as:

ω＝2πf；f_a为信号的频率均值；F_w为信号的等效频宽。In formula (2): E _p is the energy of the signal,

ω=2πf; f _a is the frequency mean of the signal; F _w is the equivalent bandwidth of the signal.

The median database of pulse current signal spectral parameters; set partial discharge sources in each component of the switchgear (part number i, i=1, 2, ..., n), obtain the pulse current signal through experiments and detection, and extract its average frequency (f _athi ), equivalent bandwidth (F _wthi ) characteristic parameters, perform k experiments at the same position, extract the frequency characteristic parameters and calculate the median value (f _athmi , F _wthmi ) of the data, the formula is:

Build a spectral feature database. As shown in Table 1, the database is an n×2 matrix.

Table 1 The median database of spectral characteristic quantities of partial discharge pulse current signal in high-voltage switchgear

(2) Calculation of spectrum parameters of switchgear pulse current signal

The time domain waveform I _ats (t) of the PD pulse current signal is measured in the switchgear. Based on the time-domain waveform, two frequency-domain characteristic quantities of the pulse current signal are calculated. The calculation formula is as formula (2). After calculation, the measured frequency-domain characteristic quantities of the switchgear pulse current are obtained: center frequency (f _ats ) and equivalent bandwidth (F _wts ).

(3) Determine the part where the partial discharge is located based on the principle of minimum Euclidean distance

Think of the database as a collection of n vectors, each vector containing 2 elements, identified by _Mi , that is:

M _i =(f _athmi ,F _wthmi ), i=1,2,...,n

The measured pulse current spectrum characteristic of the switchgear is also regarded as a vector, which is represented by S, namely:

S=(f _ats ,F _wts )

Calculate the Euclidean distance between S and _Mi one by one through the Euclidean distance formula, and select the point with the smallest Euclidean distance as the positioning result. The calculation method of Euclidean distance is as follows:

In addition, the present invention has high positioning accuracy, and can accurately determine the part where the partial discharge defect is located; and the positioning calculation has high robustness and is less affected by detection and calculation.

Claims (3)

1. The partial discharge localization method for switchgear components based on pulse current signal spectrum, is characterized in that, described method comprises the following steps: Step 1: Carry out the partial discharge test and measurement of the switchgear, establish the database of the test frequency average value f _ath and the test equivalent bandwidth F _wth of the pulse current signal, and calculate the data median value of the partial discharge of each component, and establish the spectral characteristic quantity value database; Step 2: perform partial discharge signal detection on site, collect partial discharge pulse current signals in the switch cabinet, and calculate the measured average frequency f _ats and the measured equivalent frequency width F _wts ; Step 3: Calculate the Euclidean distance D _i between the partial discharge pulse current signal collected in step 2 and the median value of the data in step 1, and determine the partial discharge component according to the principle of minimum distance. 2. The method for locating partial discharge of switchgear components based on pulse current signal spectrum according to claim 1, wherein the step 1 specifically comprises the following sub-steps: Step 11: Connect the coupling capacitor in parallel in the switch cabinet, and connect the pulse current coupling unit to its ground terminal to couple the pulse current signal; then carry out partial discharge tests on each component in the switch cabinet, and measure the pulse current signal time domain waveform I(t ); Step 12: Perform Fourier transform on the pulse current signal time-domain waveform I(t) in step 11 to obtain the pulse current signal spectrum waveform I(jω); Step 13: Extract the test frequency mean value f _ath and the test equivalent frequency width F _wth through the pulse current signal spectrum waveform I(jω), and perform multiple tests at the same position, calculate the median value of the data, and establish a median value database of spectral feature quantities . 3. The method for locating partial discharge of switchgear components based on pulse current signal spectrum according to claim 1, is characterized in that, described step 3 is as follows: The spectral feature value median database is formed into a set consisting of n vectors M _i ; the vector M _i is: M _i =(f _athmi ,F _wthmi ), i=1,2,...,n; Among them, f _athmi is the mean test frequency of the ith component, and F _wthmi is the test equivalent bandwidth of the ith component; The measured frequency mean f _ats and the measured equivalent bandwidth F _wts form a vector S, namely: S=(f _ats ,F _wts ) Calculate the Euclidean distance Di between S and _Mi one by one through the Euclidean distance formula, and select the component with the smallest Euclidean distance as the partial discharge component; the calculation method of the Euclidean distance D _i is _:

Images