CN108490311A - Weak impact signal extraction based on power frequency sampling and separation method - Google Patents

Abstract

The invention provides a weak impact signal extraction and separation method based on power frequency sampling, which includes: collecting the line signal through a high-speed sampling module, processing the sampling signal with an over-sampling algorithm, changing the sampling frequency and precision of the signal, and separating the over-sampling signal by starting the filter. The transient signal in the sampling signal is processed by the peak integrator to obtain the maximum value of the transient signal modulus, and the signal change is compared. The startup algorithm makes a judgment according to the signal change and gives a corresponding response. The present invention can effectively extract the transient signal, and the over-sampling algorithm expands the dynamic range of the startup, so that the dynamic range of the transient signal exceeds 60dB. The equipment has a high response sensitivity to transient signals, and can respond to transient signals with a power frequency signal amplitude of 0.1%. The equipment has high operating stability and strong anti-interference ability.

Description

Extraction and Separation Method of Weak Shock Signal Based on Power Frequency Sampling

technical field

The invention relates to the technical field of power equipment, in particular to power fault ranging equipment, and in particular to a method for extracting and separating weak impact signals based on power frequency sampling.

Background technique

There are two main methods of fault location in power systems: fault analysis method and traveling wave location. In recent years, the traveling wave method is more popular. The traveling wave method is a new method to measure the distance according to the propagation time of the high-frequency transient traveling wave signal generated during the fault. In practice, the length of the traveling wave propagation line and the size of the transition resistance will affect the shape and amplitude of the traveling wave head. For traveling wave ranging, the more seriously the traveling wave head is affected, the more difficult it is to identify the traveling wave head, especially noise interference In severe cases, misjudgment is likely to occur, which affects the accuracy of waveform analysis. How to effectively extract traveling waves is the prerequisite for traveling wave fault location. The energy of the traveling wave signal is much smaller than that of the power frequency signal. Only when the amplitude threshold is as low as possible can the traveling wave signal be responded to. When the amplitude threshold is low and the interference signal is relatively large, false alarms are prone to occur. The ranging equipment requires the starting weak signal to be as small as possible and has strong anti-interference ability.

Contents of the invention

In order to better respond to transient signals, improve anti-interference ability, and ensure stable and reliable operation of equipment, the present invention proposes a weak impact signal extraction and separation method based on power frequency sampling.

The present invention adopts following technical scheme:

The method for extracting and separating weak impact signals based on power frequency sampling includes the following steps:

Step 1. Obtain the voltage or current signal of the line under test;

Step 2, performing oversampling processing on the obtained voltage or current signal;

Step 3, filtering out the industrial frequency signal contained in the oversampling signal in step 2 through a digital startup filter, and extracting the transient signal in the oversampling signal;

Step 4, performing peak processing on the transient signal;

Step 5. By comparing the strength changes of the current sampling point signal and the reference sampling point signal, the type of the signal is determined, and the extraction and separation of weak impact signals based on power frequency sampling is realized.

Further, in the first step, obtaining the measured line voltage or current signal includes: power frequency signal, traveling wave signal and noise signal.

Further, in the second step, oversampling is performed on the acquired voltage or current sampling signal, and a low-frequency oversampling signal with a high number of sampling bits is obtained through an oversampling algorithm.

Further, in the step 3, the transient signal in the oversampled signal is extracted, and the specific method is: process the low-frequency oversampled signal through a digital startup filter, filter out the power frequency signal in the signal, and leave Transient signals of signal and noise signals.

Further, in step 4, peak processing is performed on the transient signal, specifically: the extracted transient signal is processed by a peak integrator to obtain the maximum modulus value of the transient signal at the current sampling moment.

Further, in the step five, by comparing the strength changes of the current sampling point signal and the reference sampling point signal, the type of the signal is determined as follows:

Judge the value of e according to algorithmic formula e=U _a -3U _(a-3) ; Wherein, U _a is current sampling point signal, and U _(a-3) is reference sampling point signal;

e≥0 is judged to be a fault traveling wave;

e<0 determines that it is an interference signal.

Further, the current sampling point signal is a sampling value closest to the current moment, and the reference sampling point is set as the fourth last sampling value.

Beneficial effects of the present invention:

1. The signal bandwidth of the sampling channel covers a wide range, which can meet the analysis requirements based on the time domain, reaching 20Hz to 1MHz.

2. The starting sensitivity is high, and the fault location device can be started when the transient signal is 0.1% of the amplitude of the power frequency signal. The operation stability of the equipment is high, and the transient signal judgment is accurate.

3. The dynamic range of the separated transient signal exceeds 60dB. The response bandwidth of the transient signal is better than 1KHz～1MHz, which meets the analysis requirements of the power system.

Description of drawings

The accompanying drawings constituting a part of the application are used to provide a further understanding of the application, and the schematic examples and descriptions of the application are used to explain the application and do not constitute a limitation to the application.

Figure 1 is a flow chart of weak impact signal extraction and separation based on power frequency sampling;

Fig. 2 is a starting filter filter attenuation curve diagram;

Figure 3(a)-(c) are the waveform diagrams of the sampled power frequency signal, step impact and its coupling signal respectively;

Fig. 4 is a signal waveform diagram processed by the startup filter;

Figure 5 is a waveform diagram of the signal processed by the peak integrator.

Detailed ways

It should be pointed out that the following detailed description is exemplary and intended to provide further explanation to the present application. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

The specific implementation will be described with reference to the flow chart in Fig. 1. In this example, the line voltage signal is sampled, and the current signal processing is similar to it. The weak impact signal adopts a step signal of V _P =500 and period = 15ms, and the power frequency signal adopts a voltage signal of V _P =20000; the sampling frequency of the high-speed A/D module is 1MHz, the number of sampling bits is 16Bit, and the sampling value is recorded as u _b ; The low-frequency oversampling frequency is 10KHz, the number of sampling bits is 23 bits, and the sampling value is recorded as U _a ;

The extraction and separation of weak impact signals based on power frequency sampling is characterized by the following steps:

Step 1: Obtain the voltage (current) signal of the line under test.

Step 2, oversampling processing of the voltage (current) sampling signal.

Step 3, extracting the transient signal in the oversampled signal.

Step 4, transient signal peak processing.

Step five, start condition analysis and judgment.

Complete the extraction and separation of weak impact signals based on power frequency sampling.

Wherein, step 1 obtains the voltage signal of the line under test. The specific method is: through the voltage sensor, the voltage signal in the power grid is collected, wherein the pass frequency band of the voltage sensor is 15Hz...5MHz. After the signal is amplified by hardware filtering, the sampling frequency is 1MHz. The high-speed A/D module is used to obtain 16-bit voltage sampling values (the sampling points are recorded as 0, 1...999999 in turn, and the sampling values are recorded as u ₀ , u ₁ ...u ₉₉₉₉₉₉ in turn). Here the sampled value u _b is a signed number. The voltage signal includes power frequency signal, traveling wave signal and noise. Figure 3(a)-(c) are waveform diagrams of power frequency signal, step impact signal and their coupling signal respectively.

Wherein, step two voltage sampling signal oversampling processing, the specific method is: the high-frequency voltage sampling signal obtained in step one is subjected to sampling processing, so that the sampling frequency of the voltage signal becomes 10KHz, and the number of sampling bits is 23 bits. The change of sampling frequency and number of bits is realized by oversampling algorithm. The algorithm is based on the formula Combine 100 consecutive high-frequency sampling values u _b into one oversampling value U _a , and the oversampling value U _a is a signed number. The voltage signal after oversampling processing still contains power frequency signal, traveling wave signal and noise.

Among them, Step 3 extracts the transient signal in the oversampled signal. The specific method is: filter the power frequency signal contained in the oversampled signal in Step 2 through a digital startup filter. The filter is designed as a single-pole filter, and the attenuation pole is 50Hz. The single-pole filter transfer function is The single-pole filter attenuates the power frequency signal by more than 80dB, and the normalized frequency is 1000Hz; its attenuation curve is shown in Figure 4. Weak impulse signals (transient signals) such as traveling wave signals and noises are obtained by attenuation.

Among them, step 4 transient signal peak value processing, the specific method is: after the digital startup filter, the power frequency signal is attenuated, the weak impact signal is retained, and the obtained weak signal is processed by the peak integrator to obtain the maximum value of the weak impact signal modulus at the current moment . The signal processed by the peak integrator is shown in Figure 4. The processed signal has only amplitude and no direction.

Among them, step five starts condition analysis and judgment, the specific method is: compare the current sampling point U _a with the reference sampling point U _(a-3) (the fourth sampling point from the bottom of the current moment) signal strength changes through a comparator, rely on the start algorithm to separate Interference signal and fault traveling wave signal, judge the signal type.

Both the current sample point and the reference sample point are signals after peak processing. The oversampling frequency is 10khz, and there are 10,000 corresponding sampling points within 1s. The current sampling point is the latest sampling value, and the reference sampling point is set to the fourth last sampling value. Assuming that there are currently 500 sampling values, the current sampling value is the 500th, and the reference sampling point is the 497th sampling value.

The value of e is judged according to the algorithm formula e=U _a -3U _(a-3) . e≥0 is considered as a fault traveling wave, and the device starts; e<0 is considered an interference signal, and the device does not start.

Although the specific implementation of the present invention has been described above in conjunction with the accompanying drawings, it does not limit the protection scope of the present invention. Those skilled in the art should understand that on the basis of the technical solution of the present invention, those skilled in the art do not need to pay creative work Various modifications or variations that can be made are still within the protection scope of the present invention.

Claims (7)

1. the weak impact signal extraction based on power frequency sampling and separation method, which is characterized in that include the following steps：

Step 1: obtaining test line voltage or current signal；

Step 2: the voltage or current signal to acquisition carry out over-sampling processing；

Step 3: starting filters filter by number falls power frequency component included in the oversampled signals of step 2, extract Transient signal in oversampled signals；

Step 4: carrying out peak value processing to transient signal；

Step 5: by comparing the Strength Changes of current sampling point signal and reference sample point signal, the type of signal is determined, it is real Weak impact signal now based on power frequency sampling is extracted and is detached.

2. the weak impact signal extraction based on power frequency sampling and separation method as described in claim 1, which is characterized in that described In step 1, obtains test line voltage or current signal includes：Power frequency component, travelling wave signal and noise signal.

3. the weak impact signal extraction based on power frequency sampling and separation method as described in claim 1, which is characterized in that described In step 2, voltage or current sampling signal to acquisition carry out over-sampling processing, and high sample bits are obtained by over-sampling algorithm Several low frequency oversampled signals.

4. the weak impact signal extraction based on power frequency sampling and separation method as described in claim 1, which is characterized in that described In step 3, the transient signal in oversampled signals is extracted, specific method is：Start filter to low frequency over-sampling by number Signal is handled, and the power frequency component in signal is filtered out, and leaves the transient signal including travelling wave signal and noise signal.

5. the weak impact signal extraction based on power frequency sampling and separation method as described in claim 1, which is characterized in that described In step 4, peak value processing is carried out to transient signal, specially：The transient signal extracted is handled by leak integrators, is obtained To current sample time transient signal mould maximum value.

6. the weak impact signal extraction based on power frequency sampling and separation method as described in claim 1, which is characterized in that described In step 5, by comparing the Strength Changes of current sampling point signal and reference sample point signal, determine that the type of signal is specific For：

According to algorithmic formula e=U_a-3U_(a-3)Judge the value of e；Wherein, U_aFor current sampling point signal, U_(a-3)For reference sample Point signal；

The judgements of e >=0 are fault traveling waves；

The judgements of e ＜ 0 are interference signals.

7. the weak impact signal extraction based on power frequency sampling and separation method as described in claim 1, which is characterized in that described Current sampling point signal is a sampled value apart from current time recently, and reference sample point is set as fourth from the last sampling Value.