CN104614586B

CN104614586B - A kind of electrical quantity signal adaptive method of sampling based on GPS

Info

Publication number: CN104614586B
Application number: CN201510016798.3A
Authority: CN
Inventors: 蒋文骏; 褚文捷; 刘亚捷; 吴颖; 王英翔
Original assignee: SHANGHAI WESCON SYSTEM CONTROL Inc
Current assignee: SHANGHAI WESCON SYSTEM CONTROL Inc
Priority date: 2015-01-13
Filing date: 2015-01-13
Publication date: 2019-01-18
Anticipated expiration: 2035-01-13
Also published as: CN104614586A

Abstract

A kind of electrical quantity signal adaptive method of sampling based on GPS, in rated frequency of the required sample rate close to chip, the degree that the dispersion of the dispersion and each sampling section width that effectively inhibit each sampling interval width increases, when making to improve sample rate in the case where the rated frequency of chip can not improve, the acquisition precision decline degree for also reducing electrical quantity is possibly realized.The present invention can predict rapidly next second clock frequency according to the variation tendency of clock frequency before, with stronger anti-interference, will not significantly it change because of the unstable generation sample frequency of pulse per second (PPS) of input, the difference especially adjusted allows more than the width of 1 sampled point, expands application range.

Description

A kind of electrical quantity signal adaptive method of sampling based on GPS

Technical field

The invention belongs to electric system observation and control technology fields, are specifically exactly to be related to a kind of electrical quantity signal based on GPS Adaptively sampled method.

Background technique

With the further expansion of electric system scale, various types of power supplys and load are linked into system, make electricity The control of Force system becomes more complicated.In order to guarantee the safe and stable operation of system, the frequency to Current Voltage, width are needed Value, harmonic wave, phase angle, the electrical quantity such as electric flux carry out high-precision measurement.In order to obtain high-precision electrical quantities measurement value, handing over It needs to control the number of interior sampled point per second not only when stream sampling to guarantee the precision of sample rate, but also needs to improve each sampling Point interval uniformity coefficient.In electrical quantity acquisition and calculating process, the sampled data needs sequence of each second and comparably distribute It is calculated to the AC sampling of number wheel, for example, need to carry out the calculating of 50 wheel electric voltage frequencies each second, sample rate 3200Hz, then 3200 point datas of each second need sequentially to evenly distribute to be calculated to this 50 wheel frequency, needed for each round calculates in other words 64 total sampling widths of sampled point, the width of referred to as sampling section is also required to as identical as possible here.Therefore in order to improve each round Calculated result precision, not only need to reduce the mutual dispersion of 3200 sampling intervals, and be also required to reduce 50 The mutual dispersion of section width is sampled, the sampling section of narrower width otherwise will be made to calculate the measured value of the a-c cycle obtained It is bigger than normal, and the measured value that the sampling section of wider width calculates the a-c cycle obtained is less than normal.

Chinese patent 201410173291.4 discloses a kind of high-precision pulse per second (PPS) based on FPGA and occurs frequently again sampling pulse Method, be denoted as method 1 below.This method compares actual synchronization pulse per second (PPS) count value and ideal synchronisation pulse per second (PPS) count value Compared with, deviation and sample frequency are subjected to accumulation calculating, judge whether pulse counter is corrected according to accumulation calculating result, from And reach and the total deviation of actual synchronization pulse per second (PPS) is evenly distributed in actual samples pulse, the error of sampling pulse is reduced, Relatively reliable stable sampling pulse signal is provided for signal acquisition.But method 1 haves the defects that several aspects, first, The deterministic process of judgement and actual samples the pulse output of accumulation amount during accumulation calculating can introduce time delay；Second, this is adopted Quadrat method needs the premise of a hypothesis: the frequency variation of crystal oscillator is very small, but the frequency of actually crystal oscillator can be with environment Temperature and device inside state consecutive variations, the sample effect that the effect of method 1 can be generated than practical crystal oscillator frequency postpone one second It appears above；Third, jamproof robustness is not strong, such as when the synchronization pulse per second (PPS) of input is unstable, such as along occurring once in a while A little earlier or a little later, that will lead to actual samples interval and also broaden together narrow, and reality is caused still to be not allowed with measurement result Really；4th, which may be only available for the factor of difference 4,8 equal 4000, and the difference adjusted can not be more than 1 sampling The width count value of point, makes application range have limitation.

In conclusion the existing electrical quantity signal sampling method scope of application is narrow, as semiconductor chip frequency reaches Physics limit, and need to further increase sample rate to obtain high resolution measurement, therefore the sample rate that will lead to is further Close to chip rated frequency when, the dispersion of sampling interval width and the dispersion for sampling section width can become larger rapidly.

Summary of the invention

The purpose of the present invention is for existing electrical quantity signal sampling method, that there are the scope of applications is narrow, measurement numerical value is still So have the defects that large error with actual value, a kind of electrical quantity signal adaptive method of sampling based on GPS is provided, is expanded The scope of application of sampling effectively inhibits point of each sampling interval width in rated frequency of the required sample rate close to chip The degree that the dispersion of scattered and each sampling section width increases, makes the raising sampling in the case where the rated frequency of chip can not improve When rate, the acquisition precision decline degree for also reducing electrical quantity is possibly realized.

Technical solution

In order to achieve the above technical purposes, the present invention designs a kind of electrical quantity signal adaptive method of sampling based on GPS, It is characterized in that, it including the following steps:

The first step receives standard pulse per second (PPS), with the clock number between the pulse per second (PPS) of clock counter continuous counter and pulse per second (PPS) It is denoted as C₀；

Second step is denoted as C using the clock number between pulse per second (PPS) recently three times₀, C_-1And C_-2, predict the second next time Clock number between pulse, is denoted as C₁；

Third step, sample rate are denoted as S=3200 point/second, and basic interval clock number is denoted as B₁=int (C₁/ S), once share equally Remainder is denoted as G₁=C₁%S, if G₁It is zero, then the sampling interval is all equal to B₁, the 6th step is jumped directly to, otherwise more than sample rate Number scale is F₁=S%G₁, into the 4th step；

4th step, if F₁It is zero, the sampled point space-number once shared equally is denoted as D₁=int (S/G₁), first sampled point A sampled point basic interval clock number in [0,3199] later is B₁, in D₁-1,2D₁-1,…,mD₁In -1 sampling interval Add the delay of 1 clock, and once sharing number scale is m=[1, int (S/D₁)], jump to the 6th step；If F₁Be not zero then into Enter the 5th step；

5th step, D₁=int (S/G₁)+1, the secondary sampled point space-number shared equally is denoted as D₂=int (S/R₂), it is secondary remaining Number scale is R₂=G₁-int(S/D₁), [0,3199] after first sampled point a sampled point basic interval clock number is B₁, In D₁-1,2D₁-1,…,mD₁Add the delay of 1 clock, and m=[1, int (S/D in -1 sampling interval₁)], in D₂-1, 2D₂-1,…,pD₂Add the delay of 1 clock in -1 sampling interval, it is secondary to share number, it is denoted as p=[1, R₂], into the 6th step；

6th step is sampled according to sample rate S according to the sampling interval of generation, jumps to second step.

Further, work as C in the second step₀, C_-1And C_-2When being initialized as 0, then C is needed_-2It is received in after actual value, It can carry out sampling action.

Further, in the second step 2, if external pulse per second (PPS) is unstable, C is calculated using equalization method₁=int ((C₀+C_-1+C_-2)/3)；C is calculated by slope if crystal oscillator often changes with temperature₁=int ((C₀+2C_-1-C_-2)/2)。

Beneficial effect

A kind of electrical quantity signal adaptive method of sampling based on GPS provided by the invention, expands the applicable model of sampling It encloses, in rated frequency of the required sample rate close to chip, effectively inhibits the dispersion and each sampling of each sampling interval width The degree that the dispersion of Duan Kuandu increases also reduces when making to improve sample rate in the case where the rated frequency of chip can not improve The acquisition precision decline degree of electrical quantity is possibly realized.The present invention can be pre- rapidly according to the variation tendency of clock frequency before Survey next second clock frequency, have stronger anti-interference, will not because of input the unstable generation sample frequency of pulse per second (PPS) it is big The variation of amplitude, the difference especially adjusted allow more than the width of 1 sampled point, expand application range.

Detailed description of the invention

Attached drawing 1 is operational flowchart of the invention.

Specific embodiment

With reference to the accompanying drawings and examples, the present invention will be further described.

Embodiment

As shown in Fig. 1, with the clock count value C of actual standard pulse per second (PPS)₀=2500020, C_-1=2500014, C_-2= 2500018, sample rate S=3200 point/second, for every wheel calculates 64 sampled points of needs,

The first step receives standard pulse per second (PPS), with the clock number between the pulse per second (PPS) of clock counter continuous counter and pulse per second (PPS) C₀, C_-1, C_-2；

Second step predicts the count value C at next pulse per second (PPS) interval₁=int ((C₀+C_-1+C_-2)/3)=2500017；

Third step, every theoretical total linear spacing for taking turns 64 sampled points in calculating cycle are C₁* 64/S=50000.34clocks； Basic interval clock number B₁=int (C₁/ S)=781；Once share remainder G equally₁=C₁%S=817；Because of G₁It is not zero, then F₁= S%G₁=749, into the 4th step,

4th step, if F₁It is not zero, into the 5th step；

5th step, D₁=int (S/G₁)+1=4, D₂=int (S/R₂)=188, R₂=G₁-int(S/D₁)=17；First 0~3199 sampled point basic interval after a sampled point is all 781 clocks；Because of D₁=4, so the 3rd, Add the delay of 1 clock, and m=1~int (S/D in the sampling interval of 7 ..., 4m-1,3199₁)；Because of D₂=188, so Add the delay of 1 clock, p=1~R in the sampling interval of 187th, 375 ..., 188p-1,3195₂。

6th step carries out 3200 point samplings according to the sampling interval of generation, jumps second step.

With 64 sampled points in every wheel calculating cycle, there is the calculating of 50 wheels in 3200 points, it is every to take turns practical total linear spacing and theory Being worth mean error is 0.0009%, and specific evaluation is as follows:

Invention which is intended to be protected is not limited only to the technical solution that above-described embodiment is related to, any by skill of the invention The enlightenment of art scheme should all fall into protection scope of the present invention.

Claims

1. a kind of electrical quantity signal adaptive method of sampling based on GPS, which is characterized in that it including the following steps:

(I) standard pulse per second (PPS) is received, is denoted as C with the clock number between the pulse per second (PPS) of clock counter continuous counter and pulse per second (PPS)₀；

(II) using the clock number between pulse per second (PPS) recently three times, it is denoted as C₀, C_-1And C_-2, predict between pulse per second (PPS) next time Clock number, be denoted as C₁；

(III) sample rate is denoted as S=3200 point/second, and basic interval clock number is denoted as B₁=int (C₁/ S), once share remainder equally, It is denoted as G₁=C₁%S, if G₁It is zero, then the sampling interval is all equal to B₁, step (VI) is jumped directly to, otherwise sample rate remainder It is denoted as F₁=S%G₁, enter step (IV)；

(IV) if F₁It is zero, the sampled point space-number once shared equally is denoted as D₁=int (S/G₁), after first sampled point [0,3199] a sampled point basic interval clock number is B₁, in D₁-1,2D₁-1,…,mD₁Add 1 in -1 sampling interval The delay of clock, and once sharing number scale is m=[1, int (S/D₁)], go to step (VI)；

If F₁It is not zero, enters step (V)；

(V)D₁=int (S/G₁)+1, the secondary sampled point space-number shared equally is denoted as D₂=int (S/R₂), quadratic remainder is denoted as R₂= G₁-int(S/D₁), [0,3199] after first sampled point a sampled point basic interval clock number is B₁, in D₁-1, 2D₁-1,…,mD₁Add the delay of 1 clock, and m=[1, int (S/D in -1 sampling interval₁)], in D₂-1,2D₂-1,…, pD₂Add the delay of 1 clock in -1 sampling interval, it is secondary to share number, it is denoted as p=[1, R₂], enter step (VI)；

(VI) it is sampled according to the sampling interval of generation according to sample rate S, go to step (II).

2. a kind of electrical quantity signal adaptive method of sampling based on GPS as described in claim 1, it is characterised in that: described Work as C in step (II)₀, C_-1And C_-2When being initialized as 0, then C is needed_-2It is received in after actual value, can just carry out sampling action.

3. a kind of electrical quantity signal adaptive method of sampling based on GPS as described in claim 1, it is characterised in that: described In step (II), if external pulse per second (PPS) is unstable, C is calculated using equalization method₁=int ((C₀+C_-1+C_-2)/3)；If Crystal oscillator often changes with temperature, calculates C by slope₁=int ((C₀+2C_-1-C_-2)/2)。