CN101762742A - Sampling value correcting method based on fixed-frequency sampling sequence - Google Patents

Abstract

The invention relates to a sampling value correcting method based on a fixed-frequency sampling sequence, which comprises the steps of data acquisition, data processing and the like. The algorithm process comprises the following concrete steps: acquiring alternate current signals in an alternate current loop according to a fixed sampling frequency of a reference frequency f, and obtaining N sampling values X1, X2...XN-1, XN within a period; measuring the actual frequency f' of the signal; and through Xk, f' and f, calculating the corrected sampling sequence X1k according to an expression, wherein k is larger than or equal to 1 and less than or equal to N.

Description

A kind of sampled value bearing calibration based on the fixed frequency sample sequence

Technical field

The present invention relates to a kind of sampled value bearing calibration, particularly a kind of sampled value bearing calibration that is applied to electric system based on the power frequency sample sequence based on the fixed frequency sample sequence.

Background technology

Mostly adopted AC Data in the power automatic device, A/D carries out analog to digital conversion with certain sample frequency by CPU control, obtain discrete sample sequence,, calculate the information such as amplitude, angle of this AC signal according to sample frequency and the sample sequence value that obtains.

In electric system, frequency is to allow to change.In A/D conversion, if weekly the N of a ripple sampled point can not be equably step by step in each power frequency period, calculate resulting AC signal result according to sample sequence and deviation will occur, cause the protective relaying device misoperation even.

Traditional synchronous sampling by software method detects cycle of AC signal in real time, adjusts sampling time interval with this, makes that weekly N sampled point of ripple is evenly distributed in each power frequency period.This method needs extra hardware circuit, and the quantization error in sampling period, the dispersiveness of interrupt response time and the measurement of signal period accurately and lag behind and also can cause sampling time error and then influence result of calculation.And when electric system was broken down, the waveform of AC signal so also can influence result of calculation because the influence of transient state process can produce very big distortion.

Approach based on linear interpolation is used in traditional sampled value bearing calibration, actual samples sampled value is constantly proofreaied and correct be the approximate sampled value in desirable sampling instant.This method is because the relation of linear interpolation, congenitally exists the bigger error of calculation, and error increases along with the increase of frequency shift (FS) power frequency.

Therefore when the AC signal frequency change, measuring this signal how simply, fast and accurately has significant meaning.

Technical scheme

The objective of the invention is in order to overcome the defective of prior art, provide a kind of simply, the method for AC signal that changes of survey frequency fast and accurately.

Technical scheme of the present invention is: a kind of sampled value bearing calibration based on the fixed frequency sample sequence, comprise data acquisition, data processing step, and this procedure specifically may further comprise the steps:

1. according to the AC signal in the reference frequency f fixed sampling frequency collection ac circuit, obtain N sampled value X in the one-period ₁, X ₂X _N-1, X _N

2. measure this signal actual frequency f ';

3. pass through X _k, f ', f, according to following formula:

(1) X ¹ _k=X _k, work as k=1;

(2) $\begin{array}{c}p=k\×\frac{50f^{\′}}{f^{\′}},r=p+1,s=p-1,\\ {X^1}_k=\frac{p}{2}\×[X_{k1}\×s+X_{k+1}\×r]X_k\×r\×s\end{array},$ As 1＜k＜N;

(3) $\begin{array}{c}p=k\×\frac{50f^{\′}}{f^{\′}},r=p+2\\ {X^1}_k=\left[\frac{p+1}{2}\right]\×[X_{k2}\×p+X_k\×r]X_{k1}\×r\×p\end{array},$ Work as k=N

Calculate revised sample sequence X ¹ _k, 1≤k≤N.

Principle of the present invention is:

Make that reference frequency is f, N is weekly the hits of phase, and actual samples at interval so $\mathrm{\Δt}=\frac{1}{\mathrm{Nf}},$

Make f ' expression signal actual frequency, so desirable sampling interval $\mathrm{\Δ}{t}^{\′}=\frac{1}{{\mathrm{Nf}}^{\′}},$ If X _{K 1}, X _k, X _K+1Be respectively k-1, k and k+1 sampled value of actual samples value sequence, X ¹ _kBe k the sampled value of sampling according to desirable sampling interval.Known X _K-1, X _k, X _K+1Three point sampling values, X ¹ _kThe sampled value of point can be approximate with quadratic interpolation, can obtain according to the quadratic interpolation formula:

(1) X ¹ _k=X _k, work as k=1;

(2) $\begin{array}{c}{X^1}_k=X_{k-1}\×\frac{[\mathrm{k\Δ}{t}^{\′}-\mathrm{k\Δt}]\×[\mathrm{k\Δ}{t}^{\′}-(k+1)\mathrm{\Δt}]}{2\mathrm{\Δ}{t}^2}\\ +X_k\×\frac{[\mathrm{k\Δ}{t}^{\′}-(k-1)\mathrm{\Δt}]\×[\mathrm{k\Δ}{t}^{\′}-(k+1)\mathrm{\Δt}]}{-\mathrm{\Δ}{t}^2}\\ +X_{k+1}\×\frac{[\mathrm{k\Δ}{t}^{\′}-(k-1)\mathrm{\Δt}]\×(\mathrm{k\Δ}{t}^{\′}-\mathrm{k\Δt})}{2\mathrm{\Δ}{t}^2}\end{array},$ As 1＜k＜N,

Carry out to obtain behind the abbreviation

${X^1}_k=\frac{X_{k-1}}{2}\×k\×\frac{f-f^{\′}}{f^{\′}}\×(k\×\frac{f}{f^{\′}}-k-1)$

$-X_k\×(k\×\frac{f}{f^{\′}}-k+1)\×(k\×\frac{f}{f^{\′}}-k-1);$

$+\frac{X_k}{2}\×(k\×\frac{f}{f^{\′}}-k+1)\×k\×\frac{f-f^{\′}}{f}$

(3) work as k=N,, can use X because sample sequence has only the N point _K-2, X _K-1, X _k3 push away X ¹ _kValue:

${X^{\′}}_k=\frac{X_{k-2}}{2}\×(k\×\frac{f}{f^{\′}}-k+1)\×k\×\frac{f-f^{\′}}{f^{\′}}$

$-X_{k-1}\×(k\×\frac{f}{f^{\′}}-k+2)\×k\×\frac{f-f^{\′}}{f^{\′}};$

$+\frac{X_k}{2}\×(k\×\frac{f}{f^{\′}}-k+2)\×(k\×\frac{f}{f^{\′}}-k+1)$

So,, can calculate revised sample sequence X according to above formula ¹ _k, 1≤k≤N, this sample sequence are evenly distributed in actual cycle of signal, just can calculate the information such as amplitude, phase place of this signal according to revised sampled value sequence.

This method is implemented simple, and physical significance is clear, on the basis that only needs fixed frequency that AC signal is sampled, for the AC signal of skew reference frequency, also can calculate the information such as amplitude phase place of AC signal fast and accurately.

Description of drawings

Fig. 1 is a method flow diagram of the present invention

Fig. 2 is that the present invention is in conjunction with the process flow diagram in protection that is used in protective relaying device and the measurement

Embodiment

Be described further below in conjunction with drawings and Examples.

The present invention is used for realizing the protection and the measurement of protective relaying device, below with the protection in the protective relaying device be measured as example and be described further:

A kind of sampled value bearing calibration based on the fixed frequency sample sequence comprises steps such as data acquisition and data processing, wherein,

Data acquisition step comprises as shown in Figure 1:

1. according to the AC signal in the reference frequency f fixed sampling frequency collection ac circuit, obtain N sampled value X in the one-period ₁, X ₂X _N-1, X _N

2. measure this signal actual frequency f ';

Data processing step comprises:

1. pass through X _k, f ' and f, according to following formula:

(1) X ¹ _k=X _k, work as k=1;

(2) $\begin{array}{c}p=k\×\frac{f-f^{\′}}{f^{\′}},r=p+1,s=p-1,\\ {X^1}_k=\frac{p}{2}\×[X_{k-1}\×s+X_{k+1}\×r]-X_k\×r\×s\end{array},$ As 1＜k＜N;

(3) $\begin{array}{c}p=k\×\frac{f-f^{\′}}{f^{\′}},r=p+2\\ {X^1}_k=\left[\frac{p+1}{2}\right]\×[X_{k-2}\×p+X_k\×r]{-X}_{k-1}\×r\×p\end{array},$ Work as k=N

Calculate revised sample sequence X ¹ _k, 1≤k≤N;

Protection in the present embodiment in the protective relaying device and measurement flow process are seen Fig. 2, comprising:

1. the initialization of protective relay, sampling module, frequency measurement module;

2. measure the AC signal actual frequency;

3. gather AC signal according to power frequency;

4. with the AC signal of gathering according to proofreading and correct based on the sampled value bearing calibration of fixed frequency down-sampling sequence;

5. calculate protection passage and the value of interchange of measuring passage according to the sample sequence after proofreading and correct, be applied to protective relay and measurement respectively.

Claims (3)

1. the sampled value bearing calibration based on the fixed frequency sample sequence comprises data acquisition and data processing step, it is characterized in that,

Described data acquisition step comprises:

1. according to the AC signal in the reference frequency f fixed sampling frequency collection ac circuit, obtain N sampled value X in the one-period ₁, X ₂... X _N-1, X _N

2. measure this signal actual frequency f ';

Described data processing step comprises:

1. pass through X _k, f ' and f, according to following formula:

(1) X ¹ _k=X _k, work as k=1;

(2) $\begin{array}{c}p=k\×\frac{f-f^{\′}}{f^{\′}},r=p+1,s=p-1,\\ {X^1}_k=\frac{p}{2}\×[X_{k-1}\×s+X_{k+1}\×r]-X_k\×r\×s\end{array},$ As 1＜k＜N;

(3) $\begin{array}{c}p=k\×\frac{f-f^{\′}}{f},r=p+2\\ {X^1}_k=\left[\frac{p+1}{2}\right]\×[X_{k-2}\×p+X_k\×r]-X_{k-1}\×r\×p\end{array},$ Work as k=N

Calculate revised sample sequence X ¹ _k, 1≤k≤N.

2. a kind of sampled value bearing calibration based on the fixed frequency sample sequence according to claim 1 is characterized in that, method of AC signal is the fixed frequency sampling in the collection ac circuit.

3. a kind of sampled value bearing calibration based on the fixed frequency sample sequence according to claim 1 is characterized in that the formula of revised sample sequence is:

(1) X ¹ _k=X _k, work as k=1;

(2) $\begin{array}{c}p=k\×\frac{f-f^{\′}}{f^{\′}},r=p+1,s=p-1,\\ {X^1}_k=\frac{p}{2}\×[X_{k-1}\×s+X_{k+1}\×r]-X_k\×r\×s\end{array},$ As 1＜k＜N;

(3) $\begin{array}{c}p=k\×\frac{f-f^{\′}}{f},r=p+2\\ {X^1}_k=\left[\frac{p+1}{2}\right]\×[X_{k-2}\×p+X_k\×r]-X_{k-1}\×r\×p\end{array}.$

Images