CN102608527A - Data filtering processing method for dynamic characteristic test for circuit breaker - Google Patents

Abstract

A data filtering processing method for a dynamic characteristic test for a circuit breaker includes steps: 1) acquiring multiple channels of voltage and current signals capable of expressing an action performance of a circuit breaker; 2) expressing voltage and current numerically via A/D (analog/digital) conversion; 3) performing hardware filtering after performing A/D (analog/digital) conversion for the acquired voltage and current signals, and uploading the voltage and current signals to an upper computer; and 4) drawing a curvilinear figure of the voltage and current signals in the upper computer, and filtering voltage and current waveforms and voltage and current values by the aid of a cubical smoothing algorithm with five-point approximation in a linear sliding average method. Waveforms filtered by cubical smoothing with five-point approximation are waveforms reflecting voltage and current. Compared with the prior art, the data filtering processing method has the following advantages that sampling precision is high, and the acquired signals can be filtered on the upper computer.

Description

Data filtering disposal route in a kind of isolating switch dynamic characteristic test

Technical field

The present invention relates to a kind of isolating switch dynamic characteristics tester, relate in particular to of the accurate processing of a kind of isolating switch dynamic characteristic test upper computer software system sampled data.

Technical background

In order to satisfy the needs of electric power development to the isolating switch performance on-line monitoring; All kinds of isolating switch testers must possess real-time monitoring operation; And the ability that can accurately measure the data of breaker operating characteristic, accurately, at a high speed, multi-channel data acquisition then is the development bottleneck with isolating switch on-line monitoring functional tester.Present isolating switch tester generally all adopts slave computer filtering, and waveform filters and is not sufficiently complete, and the subregion is bad near the hardware filtering effect, can not reflect by the time of day of sampled signal.

Summary of the invention

The objective of the invention is to overcome the deficiency of prior art, develop a kind of good stability, perfect in shape and function, adaptable isolating switch dynamic characteristic test upper computer software system

Data filtering disposal route in a kind of isolating switch dynamic characteristic test, step comprises:

1) plurality of voltages, the current signal that can represent the isolating switch performance are gathered;

2) show the form of electric current and voltage through the A/D conversion with numerical value;

3) with the voltage that collects, current signal after the A/D conversion, carry out hardware filtering, be uploaded to again in the host computer; It is characterized in that also comprising step:

4) in host computer, the draw curvilinear figure of voltage, current signal, and 53 smoothing algorithms in the applicable line moving average method carry out Filtering Processing to voltage, current waveform to voltage, current value;

In the step 1), plurality of voltages, current signal involving vibrations signal, electrical apparatus release voltage and current signal, the voltage and current signal of permanent magnetism coil, the voltage signal of linear displacement transducer.

In host computer, through the draw curvilinear figure of each channel voltage, current signal of the mode of software programming, horizontal ordinate is the time, and unit is ms; Ordinate is voltage or electric current, and V, mV or A, mA select according to the amplitude size in unit.

The linear slide method of average mainly comes this point is carried out the wave amplitude correction according to the wave amplitude of the contiguous sampled point of certain point, thereby reaches the purpose of waveform being carried out denoising, can do simply on average also can carry out weighted mean to neighbor point as required to neighbor point.Get the data point of 5 vicinities, the basic calculating formula of the method for average is:

$y_i=\underset{h=-N}{\overset{N}{\mathrm{\Σ}}}{h}_n{x}_{i-n}(i=\mathrm{1,2},...,n)$

In the formula: x is a sampled data; Y is the data after the smoothing processing; M is a number of data points; N is on average counting; H is the weighted mean factor.

The weighted mean factor must satisfy following formula:

$\underset{h=-N}{\overset{N}{\mathrm{\Σ}}}{h}_n=1$

53 times smoothly is to utilize principle of least square method that discrete data is carried out the level and smooth method of least square polynomial expression three times.

The computing formula of 5 triple smoothings is:

$\left\{\begin{array}{c}{y}_1=\frac{1}{70}[{69x}_1+4(x_2+x_4)-{6x}_3-x_5]\\ y_2=\frac{1}{35}[2(x_1+x_5)+{27x}_2+{12x}_3-{8x}_4]\\ y_i=\frac{1}{35}[-3(x_{i-2}+x_{i+2})+12(x_{i-1}+x_{i+1})+{17x}_i]\\ y_{m-1}=\frac{1}{35}[2(x_{m-4}+x_m)-{8x}_{m-3}+{12x}_{m-2}+{27x}_{m-1}]\\ y_m=\frac{1}{70}[{-x}_{m-4}+4(x_{m-3}+x_{m-1}-{6x}_{m-2})+{69x}_m]\end{array}\right.$

In the formula, i=3,4 ..., m-2.

Through the waveform empirical tests of 53 smothing filterings waveform of reflecting voltage and electric current preferably, effectively remove noise.

Smoothly on the basis of equidistant numerical value, data are handled for 53 times, established the function that y is x, y is expanded into the power series form,, get preceding four, that is: according to accuracy requirement with Taylor's formula

$y_i\left(x_i\right)=a_0+a_1{x}_i+a_2{x}_i^2+a_3{x}_i^3---\left(1\right)$

Its variance and be:

$F(a_0,a_1,a_2,a_3)=\underset{i=1}{\overset{5}{\mathrm{\Σ}}}{[(a_0+a_1{x}_i+a_2{x}_i^2+a_3{x}_i^3)-y_i]}^2---\left(2\right)$

According to principle of least square method, require (2) formula minimalization, that is:

$\begin{array}{cc}\frac{\∂F}{{\∂a}_k}=0,& k=\mathrm{0,1,2,3}\end{array}---\left(3\right)$

Following formula is launched and can obtain:

$\left\{\begin{array}{c}{5a}_0+a_1\mathrm{\Σ}{x}_j+a_2\mathrm{\Σ}{x}_j^2+a_3\mathrm{\Σ}{x}_j^3=\mathrm{\Σ}{y}_j\\ a_0\mathrm{\Σ}{x}_j+a_1\mathrm{\Σ}{x_j}^2+a_2\mathrm{\Σ}{x}_j^3+a_3\mathrm{\Σ}{x}_j^4=\mathrm{\Σ}{x}_j{y}_j\\ a_0\mathrm{\Σ}{x}_j^2+a_1\mathrm{\Σ}{x_j}^3+a_2\mathrm{\Σ}{x}_j^4+a_3\mathrm{\Σ}{x}_j^5=\mathrm{\Σ}{x}_j^2{y}_j\\ a_0\mathrm{\Σ}{x}_j^3+a_1\mathrm{\Σ}{x_j}^4+a_2\mathrm{\Σ}{x}_j^5+a_3\mathrm{\Σ}{x}_j^6=\mathrm{\Σ}{x}_j^3{y}_j\end{array}\right.---\left(4\right)$

Wherein: j=i-2, i-1, i, i+1, i+2

Because be equidistant data, get x _i=0, then corresponding 5 x value is respectively-2 Δ x ,-Δ x, and 0, Δ x, 2 Δ x then obtain following relational expression:

$\left\{\begin{array}{c}\mathrm{\Σ}{x}_j=\mathrm{\Σ}{x_j}^3=\mathrm{\Σ}{x_j}^5=0\\ \mathrm{\Σ}{x_j}^2=10{\mathrm{\Δx}}^2\\ \mathrm{\Σ}{x_j}^4=34{\mathrm{\Δx}}^4\\ \mathrm{\Σ}{x_j}^6=130{\mathrm{\Δx}}^6\\ \mathrm{\Σ}{y}_i=y_{i-2}+y_{i-1}+y_i{+y}_{i+1}+y_{i+2}\\ \mathrm{\Σ}{x}_j{y}_i=({-2y}_{i-2}-y_{i-1}+y_i+{2y}_{i+2})\mathrm{\Δx}\\ \mathrm{\Σ}{x_j}^2{y}_i=({4y}_{i-2}+y_{i-1}+y_{i+1}+{4y}_{i+2})\mathrm{\Δ}{x}^2\\ \mathrm{\Σ}{x_j}^3{y}_i=({-8y}_{i-2}-y_{i-1}+y_{i+1}+{8y}_{i+2}){\mathrm{\Δx}}^3\end{array}\right.---\left(5\right)$

(5) formula substitution (4) formula can be obtained following system of equations:

$\left\{\begin{array}{c}{5a}_0+{10a}_2{\mathrm{\Δx}}^2=y_{i-2}+y_{i-1}+y_i+y_{i+1}+y_{i+2}\\ {10a}_1\mathrm{\Δx}+34a_3{\mathrm{\Δx}}^3={-2y}_{i-2}-y_{i-1}+y_{i+1}+{2y}_{i+2}\\ {10a}_0+{34a}_2\mathrm{\Δ}{x}^2={4y}_{i-2}+y_{i-1}+y_{i+1}+{4y}_{i+2}\\ {34a}_1\mathrm{\Δx}+{130a}_3\mathrm{\Δ}{x}^3={-8y}_{i-2}-y_{i-1}+y_{i+1}+{8y}_{i+2}\end{array}\right.---\left(6\right)$

We solve a from system of equations (6) ₀, a ₁, a ₂, a ₃Value be respectively:

$\left\{\begin{array}{c}{a}_0=\frac{1}{35}({-3y}_{i-2}+{12y}_{i-1}+{17y}_i+{12y}_{i+1}-{3y}_{i+2})\\ a_1=\frac{1}{10\mathrm{\Δx}}(\frac{5}{6}{y}_{i-2}-\frac{20}{3}{y}_{i-1}+\frac{20}{3}{y}_{i+1}-\frac{5}{6}{y}_{i+2})\\ a_2=\frac{1}{14{\mathrm{\Δx}}^2}({2y}_{i-2}-y_{i-1}-{2y}_i-y_{i+1}+{2y}_{i+2})\\ a_3=\frac{1}{12{\mathrm{\Δx}}^3}({-y}_{i-2}+y_{i-1}-{2y}_{i+1}+y_{i+2})\end{array}\right.---\left(7\right)$

Again with the value-2 Δ x of x ,-Δ x, 0, Δ x, 2 Δ x substitution formulas (1) respectively draw:

$\left\{\begin{array}{c}{y}_{i-2}=a_0-{2a}_1\mathrm{\Δx}+{4a}_2{\mathrm{\Δx}}^2-{8a}_3{\mathrm{\Δx}}^3\\ y_{i-1}=a_0-{2a}_1\mathrm{\Δx}+{4a}_2\mathrm{\Δ}{x}^2-{8a}_3{\mathrm{\Δx}}^3\\ y_i=a_0\\ y_{i+1}=a_0+a_1\mathrm{\Δx}+a_2\mathrm{\Δ}{x}^2+a_3{\mathrm{\Δx}}^3\\ y_{i+2}=a_0+{2a}_1\mathrm{\Δx}+{4a}_2{\mathrm{\Δx}}^2+{8a}_3{\mathrm{\Δx}}^3\end{array}\right.---\left(8\right)$

(7) formula substitution (8) formula, the final expression formula that obtains 53 methods is at last:

$\left\{\begin{array}{c}{\stackrel{\‾}{y}}_{i-2}=\frac{1}{70}({69y}_{i-2}+{4y}_{i-1}-{6y}_i+{4y}_{i+1}-y_{i+2})\\ {\stackrel{\‾}{y}}_{i-1}=\frac{1}{35}({2y}_{i-2}+{27y}_{i-1}+{12y}_i-{8y}_{i+1}+{2y}_{i+2})\\ {\stackrel{\‾}{y}}_i=\frac{1}{35}({-3y}_{i-2}+{12y}_{i-1}+{17y}_i+{12y}_{i+1}-{3y}_{i+2})\\ {\stackrel{\‾}{y}}_{i+1}=\frac{1}{35}({2y}_{i-2}-{8y}_{i-1}+{12y}_i+{27y}_{i+1}+{2y}_{i+2})\\ {\stackrel{\‾}{y}}_{i+2}=\frac{1}{70}({-y}_{i-2}+{4y}_{i-1}-{6y}_i+{4y}_{i+1}+{69y}_{i+2})\end{array}\right.$

Compare with prior art, the present invention has following good effect or advantage:

1, sampling precision is high, can reach 40kHz, can be stable collect voltage of signals, current value, and reflect data variation accurately.

2,53 smoothing algorithms are compiled in the VC++ host computer, can realize the filtering of acquired signal at host computer, if the combined with hardware circuit filtering, the waveform situation of the signal that can extraordinary reflection collects.

Description of drawings

Fig. 1 journey software flow pattern of serving as theme.

Embodiment

For more well-known explanation the object of the invention, technical scheme and advantage thereof, the present invention is further specified below in conjunction with accompanying drawing.Specific embodiment described herein is only in order to explaining the present invention, and is not used in qualification the present invention.

Below in conjunction with description of drawings and embodiment the present invention is explained further details:

In the time of practical implementation of the present invention,

1) use DSP28335 and combine the external hardware circuit that plurality of voltages, the current signal that can represent the isolating switch performance carried out high precision collecting,

2) show the form of electric current and voltage through the A/D conversion, can gather vibration signal, electrical apparatus release voltage and current signal, the voltage and current signal of permanent magnetism coil, the voltage signal of linear displacement transducer etc. with numerical value.

3) with the voltage and current signal that collects after A/D conversion, carry out hardware filtering, be uploaded in the VC++ host computer,

4) through the draw curvilinear figure of voltage, current signal of VC++, and 53 smoothing algorithms in the applicable line moving average method carry out Filtering Processing to voltage current waveform, remove the noise in the signal in conjunction with the external hardware circuit.

As shown in Figure 1, the process flow diagram of software systems signals collecting of the present invention and processing main thread:

At first, after host computer sends data acquisition instruction, slave computer will carry out the A/D conversion to acquisition target, be digital signal with the analog signal conversion of voltage, electric current.Digital signal gets into the digital signal acquiring stage after being prone to the noise signal of identification through the hardware circuit filtering.The digital signal samples frequency is 40kHz, and precision is high, whole states of reflection digital signal that can be complete.Next, slave computer will be uploaded to the digital signal that collects in the host computer through data communication module.After host computer receives data, sampled data is carried out the smoothed data Filtering Processing 53 times, remove the effectively noise signal of removal of failing in the slave computer sample circuit.Host computer shows smooth curve that can true reflected signal state, and is saved to database after the Filtering Processing of accomplishing signal, waits until when needing and calls.

Claims (3)

1. the data filtering disposal route in the isolating switch dynamic characteristic test, step comprises:

1) plurality of voltages, the current signal that can represent the isolating switch performance are gathered;

2) show the form of electric current and voltage through the A/D conversion with numerical value;

3) with the voltage that collects, current signal after the A/D conversion, carry out hardware filtering, be uploaded to again in the host computer; It is characterized in that also comprising step:

4) in host computer, the draw curvilinear figure of voltage, current signal, and 53 smoothing algorithms in the applicable line moving average method carry out Filtering Processing to voltage, current waveform to voltage, current value;

The basic calculating formula of the linear slide method of average is: $y_i=\underset{h=-N}{\overset{N}{\mathrm{\Σ}}}{h}_n{x}_{i-n}(i=\mathrm{1,2},...,n),$

In the formula: x is a sampled data; Y is the data after the smoothing processing; M is a number of data points; N is the slip order; H is the weighted mean factor; The weighted mean factor must satisfy following formula:

$\underset{h=-N}{\overset{N}{\mathrm{\Σ}}}{h}_n=1$

Then get the data point of 5 vicinities, N gets 3, carries out the calculating of 5 triple smoothings, and formula is:

$\left\{\begin{array}{c}{y}_1=\frac{1}{70}[{69x}_1+4(x_2+x_4)-{6x}_3-x_5]\\ y_2=\frac{1}{35}[2(x_1+x_5)+{27x}_2+{12x}_3-{8x}_4]\\ y_i=\frac{1}{35}[-3(x_{i-2}+x_{i+2})+12(x_{i-1}+x_{i+1})+{17x}_i]\\ y_{m-1}=\frac{1}{35}[2(x_{m-4}+x_m)-{8x}_{m-3}+{12x}_{m-2}+{27x}_{m-1}]\\ y_m=\frac{1}{70}[{-x}_{m-4}+4(x_{m-3}+x_{m-1}-{6x}_{m-2})+{69x}_m]\end{array}\right.$

In the formula, i=3,4 ..., m-2;

Waveform through 53 smothing filterings is the waveform of reflecting voltage and electric current.

2. the data filtering disposal route in the isolating switch dynamic characteristic test according to claim 1; It is characterized in that; In the step 1), plurality of voltages, current signal involving vibrations signal, electrical apparatus release voltage and current signal, the voltage and current signal of permanent magnetism coil, the voltage signal of linear displacement transducer.

3. the data filtering disposal route in the isolating switch dynamic characteristic test according to claim 1 is characterized in that in host computer, and through the draw curvilinear figure of each channel voltage, current signal of the mode of software programming, horizontal ordinate is the time, and unit is ms; Ordinate is voltage or electric current, and V, mV or A, mA select according to the amplitude size in unit.

Images