CN104090159A - Electric energy measuring method and device - Google Patents

Abstract

The invention provides an electric energy measuring method and device. The method includes the step of conducting low-pass filtering on electric signals so as to obtain analog signals, the step of conducting A/D sampling on the analog signals so as to obtain multiple discrete signals, the step of conducting improved generalized rapid S conversion on each discrete signal so as to obtain a first result matrix corresponding to the discrete signal, the step of conducting linear decomposition on each first result matrix through the linear characteristic of the improved generalized rapid S conversion so as to obtain a second result matrix corresponding to each frequency component of each discrete signal, the step of conducting inverse improved generalized rapid S conversion on each second result matrix according to the inverse nondestructive characteristic so as to obtain a reconstructed time domain signal corresponding to each frequency component of each discrete signal, and the step of obtaining the electric energy consumption value within the A/D sampling time interval so as to obtain the electric energy consumption value within the preset time. By means of the electric energy measuring method and device, the Gauss window shape can be adjusted according to the actual condition, and operation efficiency is improved.

Description

Electric energy gauging method and device

Technical field

The application relates to power technology field, particularly a kind of electric energy gauging method and device.

Background technology

Electric energy be modern society produce and life in indispensable important energy source commodity, it produce, sell and use the system being formed by generating plant, power pack and user tripartite that depends on.Therefore, the accuracy of electric energy metrical and rationality directly have influence on send out, for, tripartite's economic interests and the fairness of transaction.

At present, domestic and international existing electric energy meter can only carry out electric energy metrical to steady-state distortion signal, and the electric energy metrical of transient state distorted signal is seemed to helpless.But input of the frequent starting of motor, load etc. all can produce huge transient power in industry spot, and electric energy meter cannot be realized metering, thereby makes power department suffer huge loss.Therefore, further investigation power network signal there is the impact on electric energy metrical of when distortion and under this background, how to realize accurately, reasonably electric energy metrical has important theory and realistic meaning.

Conventional electric energy metrical algorithm has S conversion (Stockwell Transform) at present.S conversion is the algorithm that R.G.Stockwell proposed and detected for seismic signal in 1996, and it is a kind of reversible Time-Frequency Analysis Method.But S converts the shortcoming that exists Gaussian window shape to adjust according to actual conditions, and because the time-frequency matrix information amount of S mapping algorithm is huge, the time that computing is expended is long, and operation efficiency is low.

Summary of the invention

For solving the problems of the technologies described above, the embodiment of the present application provides a kind of electric energy gauging method and device, realizes Gaussian window shape and can adjust according to actual conditions, and improve the object of operation efficiency to reach, and technical scheme is as follows:

A kind of electric energy gauging method, comprising:

Carry out low-pass filtering to carrying out electric signal in the electrical network of electric energy metrical, obtain simulating signal;

Described simulating signal is carried out to A/D sampling, obtain multiple discrete signals;

Respectively various discrete signal is improved to the quick S conversion of broad sense, obtain each self-corresponding the first matrix of consequence of various discrete signal, the quick S conversion of described improvement broad sense is on the basis of S conversion, introduce the Gaussian window width adjusting factor and the DC component place Gaussian window width adjusting factor and adopt the think of of fast algorithm conceivable;

Utilize the linear characteristic of improving the quick S conversion of broad sense, each first matrix of consequence is carried out to linearity and decompose, obtain second matrix of consequence corresponding to each frequency component of various discrete signal;

Each second matrix of consequence is carried out to the contrary quick S conversion of broad sense that improves according to inversion non-destructive respectively, obtain the reconstruct time-domain signal of each frequency component of various discrete signal;

According to electric energy metrical requirement, each reconstruct time-domain signal is carried out to corresponding point multiply each other and result of product is cumulative, the result of product accumulated value of each reconstruct time-domain signal is added and obtains cumulative sum, described cumulative sum is multiplied by the A/D sampling interval time, obtain the power consumption value within the described A/D sampling interval time, to obtain the power consumption value in Preset Time.

Preferably, respectively various discrete signal is improved to the quick S conversion of broad sense, obtains the process of each self-corresponding the first matrix of consequence of various discrete signal, comprising:

Adopt the quick S conversion of discrete improvement broad sense expression formula $\left\{\begin{array}{c}S[a,n]=\underset{p=-{\mathrm{width}}_n\left(n\right)}{\overset{{\mathrm{width}}_n\left(n\right)-1}{\mathrm{\Σ}}}H[p+{\mathrm{centre}}_n\left(n\right)]e^{-\frac{{2\mathrm{\π}}^2{(p+\frac{\mathrm{N\γ}}{f_s})}^{2\mathrm{\β}}{\mathrm{\α}}^2}{{{\mathrm{centre}}_v}^2\left(n\right)}}{e}^{j\frac{2\mathrm{\πpa}}{N}}\\ S[a,0]=\frac{1}{N}\underset{K=0}{\overset{N-1}{\mathrm{\Σ}}}h\left(\mathrm{k\Δt}\right)\end{array}\right.$ Various discrete signal is improved to the quick S conversion of broad sense, obtain each self-corresponding the first matrix of consequence of various discrete signal, described in $S[a,n]=\underset{p=-{\mathrm{width}}_n\left(n\right)}{\overset{{\mathrm{width}}_n\left(n\right)-1}{\mathrm{\Σ}}}H[p+{\mathrm{centre}}_n\left(n\right)]e^{-\frac{{2\mathrm{\π}}^2{(p+\frac{\mathrm{N\γ}}{f_s})}^{2\mathrm{\β}}{\mathrm{\α}}^2}{{{\mathrm{centre}}_v}^2\left(n\right)}}{e}^{j\frac{2\mathrm{\πpa}}{N}}$ For the quick S conversion of the discrete improvement broad sense expression formula of AC portion, described in for the quick S conversion of the discrete improvement broad sense expression formula of direct current component, a=0,1 ... N-1, n=1 ... n _max-1, p=-width (n) ..., 0 ... width (n)-1, the sampling number that described N is electric signal to be analyzed, the sampling time interval that described Δ t is electric signal to be analyzed, (k Δ is t) sampled value of each sampling instant to described h.

Preferably, the quick S conversion of described discrete improvement broad sense expression formula obtains by the quick S transformation for mula of improvement broad sense is carried out to discrete transform, and the quick S transformation for mula of described improvement broad sense is $S(t,f)={\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)w_m(t-\mathrm{\τ},\mathrm{\α},\mathrm{\β},\mathrm{\γ})e^{-j2\mathrm{\πf\τ}}\mathrm{d\τ},$ Described $w_m(t-\mathrm{\ι},\mathrm{\α},\mathrm{\β},\mathrm{\γ})=\frac{\mathrm{\α}{\left(\right|f|+\mathrm{\γ})}^{\mathrm{\β}}}{\sqrt{2\mathrm{\π}}}{e}^{-\frac{{\mathrm{\α}}^2{\left(\right|f|+\mathrm{\γ})}^{2\mathrm{\β}}{(t-\mathrm{\ι})}^2}{2}},$ H (τ) is electric signal to be analyzed, and t is time shift variable, and f is frequency, and τ is the time, and α and β are the Gaussian window width adjusting factors, and γ is the DC component place Gaussian window width adjusting factor, w _m(t-τ, α, beta, gamma) is to improve Generalized Gaussian window;

The process of wherein, carrying out discrete transform to improving broad sense quick S transformation for mula is:

A: utilize formula calculate the frequency domain radius of described improvement Generalized Gaussian window, described in for described frequency domain radius;

B: the frequency values of calculated rate sampled point number and center frequency points, described frequency sampling is counted out and is designated as n _max, the frequency values of described center frequency points is designated as f _n, described center frequency points is frequency sampling point;

C: the discrete radius of improvement Generalized Gaussian window that calculates each center frequency points is described is width _n(n) improve the discrete radius of Generalized Gaussian window, to round symbol;

D: the shift length that calculates the frequency spectrum of each center frequency points is described centre _nfor shift length;

E: calculate the coverage of each frequency domain of described improvement Generalized Gaussian window, coverage is by f _startand f (n) _end(n) characterize, wherein, f _start(0)=0, f _start(n+1)=centre _n(n)+width _n(n)+width _n(n+1), n=0 ..., n _max-1, f _end(n)=f _start(n)+2width _n(n);

F: the fast Fourier transform FFT result of calculating described electric signal to be analyzed, and described FFT result is shifted, obtain a FFT result, be designated as H[p], direct current result in a described FFT result intermediate frequency spectrum information is positioned at center, and negative frequency component and positive frequency component lay respectively at both sides;

G: the time domain radius that calculates the improvement Generalized Gaussian window of each center frequency points is $W_m[p,\mathrm{centre}\left(n\right)]=\exp [-\frac{2{\mathrm{\π}}^2{(p+\frac{\mathrm{N\γ}}{f_s})}^{2\mathrm{\β}}{\mathrm{\α}}^2}{{{\mathrm{centre}}_n}^2\left(n\right)}],$ p＝-width _n(n)，-width _n(n)+1，...，0，...，width _n(n)-1；

H: a described FFT result is added with the shift length of the frequency spectrum of each center frequency points respectively, obtains each self-corresponding the 2nd FFT result of each center frequency points, be designated as H[p+centre _n(n)];

I: the 2nd FFT result separately of each center frequency points and the time domain radius of its improvement Generalized Gaussian window are separately multiplied each other, obtain the operation result of each center frequency points, each operation result is arranged according to distributing position in FFT frequency spectrum, obtain rank results, described rank results is carried out to IFFT operation, obtain the quick S conversion of the discrete improvement broad sense expression formula of AC portion.

Preferably, described step B comprises:

B11: the frequency that makes n frequency sampling point is f _n(n is since 0), the relation of adjacent two centre frequencies: $f_n+\frac{\mathrm{\α}{\left(\right|f_n|+\mathrm{\γ})}^{\mathrm{\β}}}{2\sqrt{2}\mathrm{\π}}+\frac{\mathrm{\α}{\left(\right|f_{n+1}|+\mathrm{\γ})}^{\mathrm{\β}}}{2\sqrt{2}\mathrm{\π}}=f_{n+1};$

B12: making first frequency central point is that dc point is f ₀=0, and foundation obtain the frequency values f of each center frequency points _n, n=0 ..., n _max-1;

B13: solve inequality obtain the frequency sampling n that counts out _max, described f _sfor the sample frequency of described electric signal to be analyzed.

Preferably, utilize the linear characteristic of improving the quick S of broad sense conversion, each first matrix of consequence carried out to linearity decomposition, obtain the process of the second matrix of consequence corresponding to each frequency component of various discrete signal, comprising:

Utilize the linear characteristic S[h (t) that improves the quick S conversion of broad sense]=S[x (t)+y (t)]=S[x (t)]+S[y (t)], in the time calculating certain frequency component of each the first matrix of consequence, only retain in described the first matrix of consequence part that should frequency component, making part assignment corresponding to the frequency component except this frequency component in described the first matrix of consequence is 0;

Described h (t), x (t), y (t) are signal to be analyzed and h (t)=x (t)+y (t).

Preferably, described inversion non-destructive is: ${\∫}_{-\∞}^{+\∞}{w}_m(t-\mathrm{\τ})\mathrm{dt}={\∫}_{-\∞}^{+\∞}\frac{1}{\mathrm{\σ}\sqrt{2\mathrm{\π}}}{e}^{-\frac{{(t-\mathrm{\τ})}^2}{{2\mathrm{\σ}}^2}}\mathrm{dt}=1,$

$\begin{array}{c}{\∫}_{-\∞}^{+\∞}S(t,f)\mathrm{dt}={\∫}_{-\∞}^{+\∞}\left[{\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)w_m(t-\mathrm{\τ})e^{-j2\mathrm{\πf\τ}}\mathrm{d\τ}\right]\mathrm{dt}={\∫}_{-\∞}^{+\∞}\left[{\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)w_m(t-\mathrm{\τ})e^{-j2\mathrm{\πf\τ}}\mathrm{dt}\right]\mathrm{d\τ}\\ ={\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)e^{-j2\mathrm{\πf\τ}}\left[{\∫}_{-\∞}^{+\∞}{w}_m(t-\mathrm{\τ})\mathrm{dt}\right]\mathrm{d\τ}={\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)e^{-j2\mathrm{\πf\τ}}\mathrm{d\τ}=H\left(f\right)\end{array},$

$h\left(\mathrm{\τ}\right)={\∫}_{-\∞}^{+\∞}H\left(f\right)e^{j2\mathrm{\πf\τ}}\mathrm{d\τ}={\∫}_{-\∞}^{+\∞}\left[{\∫}_{-\∞}^{+\∞}S(t,f)\mathrm{dt}\right]e^{j2\mathrm{\πf\π}}\mathrm{d\τ};$

The described contrary principle of improving the quick S conversion of broad sense is

$\begin{array}{c}\mathrm{\α}(v,f)={\∫}_{-\∞}^{+\∞}S(t,f)e^{-j2\mathrm{\πvt}}\mathrm{dt}=\\ {\∫}_{-\∞}^{+\∞}\left[{\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)w_m(t-\mathrm{\τ})e^{-j2\mathrm{\πf\τ}}\mathrm{d\τ}\right]e^{-j2\mathrm{\πvt}}\mathrm{dt}=\\ {\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)\left[{\∫}_{-\∞}^{+\∞}{w}_m(t-\mathrm{\τ})e^{-j2\mathrm{\πvt}}\mathrm{dt}\right]e^{-j2\mathrm{\πft}}\mathrm{dt}=\\ {\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)\left[{\∫}_{-\∞}^{+\∞}{w}_m\left(t\right)e^{-j2\mathrm{\πv}(t+\mathrm{\τ})}\mathrm{dt}\right]e^{-j2\mathrm{\πf\τ}}\mathrm{d\τ}=\\ {\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)\left[{\∫}_{-\∞}^{+\∞}{w}_m\left(t\right)e^{-j2\mathrm{\πvt}}\mathrm{dt}\right]e^{-j2\mathrm{\πv\τ}}{e}^{-j2\mathrm{\πf\τ}}\mathrm{d\τ}=\\ {\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)W_m\left(v\right)e^{-j2\mathrm{\πv\τ}}{e}^{-j2\mathrm{\πf\τ}}\mathrm{d\τ}=H(v+f)W_m\left(v\right)\end{array},$

H(v+f)＝α(v，f)/W _m(v)，

Described a (v, f) be the Fourier transform result to described time shift variable t and the Fourier transform result to described time τ, described v is the Fourier transform to described time shift variable t, described f is the Fourier transform to described τ, and H (v+f) is the Fourier transform result to electric signal to be analyzed.

Preferably, according to electric energy metrical requirement, any one reconstruct time-domain signal is carried out that corresponding point multiply each other and by process cumulative result of product, being comprised:

This reconstruct time-domain signal is separated, obtain each harmonic voltage, electric current discrete signal, be designated as respectively u _n[k], i _n[k];

According to formula calculate the electric energy of this reconstruct time-domain signal unit interval internal consumption.

A kind of electric power meter, comprising:

Filtration module, for the electrical network electric signal that need to carry out electric energy metrical is carried out to low-pass filtering, obtains simulating signal;

Sampling module, for described simulating signal is carried out to A/D sampling, obtains multiple discrete signals;

The first conversion module, for respectively various discrete signal being improved to the quick S conversion of broad sense, obtain each self-corresponding the first matrix of consequence of various discrete signal, the quick S conversion of described improvement broad sense is on the basis of S conversion, introduce the Gaussian window width adjusting factor and the DC component place Gaussian window width adjusting factor and adopt the think of of fast algorithm conceivable;

Decomposing module, for utilizing the linear characteristic of improving the quick S conversion of broad sense, carries out linearity to each first matrix of consequence and decomposes, and obtains second matrix of consequence corresponding to each frequency component of various discrete signal;

The second conversion module, for each second matrix of consequence is carried out to the contrary quick S conversion of broad sense that improves according to inversion non-destructive respectively, obtains the reconstruct time-domain signal of each frequency component of various discrete signal;

Computing module, be used for according to electric energy metrical requirement, each reconstruct time-domain signal is carried out to corresponding point multiply each other and result of product is cumulative, the result of product accumulated value of each reconstruct time-domain signal is added and obtains cumulative sum, described cumulative sum is multiplied by the A/D sampling interval time, obtain the power consumption value within the described A/D sampling interval time, to obtain the power consumption value in Preset Time.

Preferably, described the first conversion module comprises:

The first converter unit, for adopting the quick S conversion of discrete improvement broad sense expression formula

$\left\{\begin{array}{c}S[a,n]=\underset{p=-{\mathrm{width}}_n\left(n\right)}{\overset{{\mathrm{width}}_n\left(n\right)-1}{\mathrm{\Σ}}}H[p+{\mathrm{centre}}_n\left(n\right)]e^{-\frac{{2\mathrm{\π}}^2{(p+\frac{\mathrm{N\γ}}{f_s})}^{2\mathrm{\β}}{\mathrm{\α}}^2}{{{\mathrm{centre}}_v}^2\left(n\right)}}{e}^{j\frac{2\mathrm{\πpa}}{N}}\\ S[a,0]=\frac{1}{N}\underset{K=0}{\overset{N-1}{\mathrm{\Σ}}}h\left(\mathrm{k\Δt}\right)\end{array}\right.$

Various discrete signal is improved to the quick S conversion of broad sense, obtain each self-corresponding the first matrix of consequence of various discrete signal, described in $S[a,n]=\underset{p=-{\mathrm{width}}_n\left(n\right)}{\overset{{\mathrm{width}}_n\left(n\right)-1}{\mathrm{\Σ}}}H[p+{\mathrm{centre}}_n\left(n\right)]e^{-\frac{{2\mathrm{\π}}^2{(p+\frac{\mathrm{N\γ}}{f_s})}^{2\mathrm{\β}}{\mathrm{\α}}^2}{{{\mathrm{centre}}_v}^2\left(n\right)}}{e}^{j\frac{2\mathrm{\πpa}}{N}}$ For the quick S conversion of the discrete improvement broad sense expression formula of AC portion, described in for the quick S conversion of the discrete improvement broad sense expression formula of direct current component, a=0,1 ... ..N-1, n=1 ... ..n _max-1, p=-width (n) ..., 0 ... width (n)-1, the sampling number that described N is electric signal to be analyzed, the sampling time interval that described Δ t is electric signal to be analyzed, (k Δ is t) sampled value of each sampling instant to described h;

The quick S conversion of described discrete improvement broad sense expression formula obtains by the quick S transformation for mula of improvement broad sense is carried out to discrete transform, and the quick S transformation for mula of described improvement broad sense is $S(t,f)={\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)w_m(t-\mathrm{\τ},\mathrm{\α},\mathrm{\β},\mathrm{\γ})e^{-j2\mathrm{\πf\τ}}\mathrm{d\τ},$ Described $w_m(t-\mathrm{\τ},\mathrm{\α},\mathrm{\β},\mathrm{\γ})=\frac{\mathrm{\α}{\left(\right|f|+\mathrm{\γ})}^{\mathrm{\β}}}{\sqrt{2\mathrm{\π}}}{e}^{-\frac{{\mathrm{\α}}^2{\left(\right|f|+\mathrm{\γ})}^{2\mathrm{\β}}{(t-\mathrm{\τ})}^2}{2}},$ H (τ) is electric signal to be analyzed, and t is time shift variable, and f is frequency, and τ is the time, and α and β are the Gaussian window width adjusting factors, and γ is the DC component place Gaussian window width adjusting factor, w _m(t-τ, α, beta, gamma) is to improve Generalized Gaussian window.

Preferably, also comprise:

The 3rd conversion module, for carrying out discrete transform to improving the quick S transformation for mula of broad sense;

Described the 3rd conversion module comprises:

The first computing unit, for utilizing formula calculate the frequency domain radius of described improvement Generalized Gaussian window, described in for described frequency domain radius;

The second computing unit, for the frequency values of calculated rate sampled point number and center frequency points, described frequency sampling is counted out and is designated as n _max, the frequency values of described center frequency points is designated as f _n, described center frequency points is frequency sampling point;

The 3rd computing unit, for the discrete radius of improvement Generalized Gaussian window that calculates each center frequency points is described is width _n(n) improve the discrete radius of Generalized Gaussian window, to round symbol;

The 4th computing unit, for the shift length of the frequency spectrum that calculates each center frequency points is described centre _nfor shift length;

The 5th computing unit, for calculating the coverage of each frequency domain of described improvement Generalized Gaussian window, coverage is by f _startand f (n) _end(n) characterize, wherein, f _start(0)=0, f _start(n+1)=centre _n(n)+width _n(n)+width _n(n+1), n=0 ..., n _max-1, f _end(n)=f _start(n)+2width _n(n);

The 6th computing unit, for calculating the FFT result of described electric signal to be analyzed, and is shifted to described FFT result, obtain a FFT result, be designated as H[p], the direct current result in a described FFT result intermediate frequency spectrum information is positioned at center, and negative frequency component and positive frequency component lay respectively at both sides;

The 7th computing unit, for the time domain radius of the improvement Generalized Gaussian window that calculates each center frequency points is $W_m[p,\mathrm{centre}\left(n\right)]=\exp [-\frac{2{\mathrm{\π}}^2{(p+\frac{\mathrm{N\γ}}{f_s})}^{2\mathrm{\β}}{\mathrm{\α}}^2}{{{\mathrm{centre}}_n}^2\left(n\right)}],$ p＝-width _n(n)，-width _n(n)+1，...，0，...，width _n(n)-1；

The 8th computing unit, for a described FFT result is added with the shift length of the frequency spectrum of each center frequency points respectively, obtains each self-corresponding the 2nd FFT result of each center frequency points, is designated as H[p+centre _n(n)];

The 9th computing unit, for the 2nd FFT result separately of each center frequency points and the time domain radius of its improvement Generalized Gaussian window are separately multiplied each other, obtain the operation result of each center frequency points, each operation result is arranged according to distributing position in FFT frequency spectrum, obtain rank results, described rank results is carried out to IFFT operation, obtain the quick S conversion of the discrete improvement broad sense expression formula of AC portion.

Preferably, described the second computing unit comprises: the first computation subunit, using the frequency of n frequency sampling point is up to the present f _n(n is since 0), the relation of adjacent two centre frequencies:

$f_n+\frac{\mathrm{\α}{\left(\right|f_n|+\mathrm{\γ})}^{\mathrm{\β}}}{2\sqrt{2}\mathrm{\π}}+\frac{\mathrm{\α}{\left(\right|f_{n+1}|+\mathrm{\γ})}^{\mathrm{\β}}}{2\sqrt{2}\mathrm{\π}}=f_{n+1};$

The second computation subunit, with first frequency central point be up to the present that dc point is f ₀=0, and foundation obtain the frequency values f of each center frequency points _n, n=0 ..., n _max-1;

The 3rd computation subunit, for solving inequality obtain the frequency sampling n that counts out _max, described f _sfor the sample frequency of described electric signal to be analyzed.

Preferably, described decomposing module is specifically for utilizing the linear characteristic S[h (t) that improves the quick S conversion of broad sense]=S[x (t)+y (t)]=S[x (t)]+S[y (t)], in the time calculating certain frequency component, only retain in described the first matrix of consequence part that should frequency component, making part assignment corresponding to the frequency component except this frequency component in described the first matrix of consequence is 0;

Described h (t), x (t), y (t) are signal to be analyzed and h (t)=x (t)+y (t).

Preferably, described computing module comprises:

Separative element, for this reconstruct time-domain signal is separated, obtains each harmonic voltage, electric current discrete signal, is designated as respectively u _n[k], i _n[k];

The tenth computing unit, for according to formula calculate the electric energy of this reconstruct time-domain signal unit interval internal consumption.

Compared with prior art, the application's beneficial effect is:

In this application, introduced the Gaussian window width adjusting factor and the DC component place Gaussian window width adjusting factor owing to improving the quick S conversion of broad sense on the basis of S conversion, therefore Gaussian window shape can be adjusted according to actual conditions.

Convert based on fast algorithm owing to improving the quick S of broad sense, therefore in the time adopting improvement generalized S-transform, improved arithmetic speed, and in conjunction with utilizing the linear characteristic of improving the quick S conversion of broad sense to carry out linearity decomposition, reduced calculated amount, improved operation efficiency.

Brief description of the drawings

In order to be illustrated more clearly in the technical scheme in the embodiment of the present application, below the accompanying drawing of required use during embodiment is described is briefly described, apparently, accompanying drawing in the following describes is only some embodiment of the application, for those of ordinary skill in the art, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.

Fig. 1 is a kind of process flow diagram of electric energy gauging method based on improving the quick S of broad sense conversion that the application provides;

Fig. 2 is a kind of structural representation of the electric power meter of the quick S conversion of the improvement broad sense that provides of the application.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present application, the technical scheme in the embodiment of the present application is clearly and completely described, obviously, described embodiment is only some embodiments of the present application, instead of whole embodiment.Based on the embodiment in the application, those of ordinary skill in the art are not making the every other embodiment obtaining under creative work prerequisite, all belong to the scope of the application's protection.

Embodiment mono-

Refer to Fig. 1, a kind of process flow diagram of electric energy gauging method based on improving the quick S conversion of broad sense that it shows that the application provides, can comprise the following steps:

Step S11: carry out low-pass filtering to carrying out electric signal in the electrical network of electric energy metrical, obtain simulating signal.

In the present embodiment, electric signal comprises current signal and voltage signal.

In the present embodiment, according to the setting of actual metered requirement and A/D sampling rate, determine the highest frequency value of the electric signal that can process, then utilize analogue filter circuit to carry out filtering processing to electric signal.

Step S12: described simulating signal is carried out to A/D sampling, obtain multiple discrete signals.

Step S13: respectively various discrete signal is improved to the quick S conversion of broad sense, obtain each self-corresponding the first matrix of consequence of various discrete signal, the quick S conversion of described improvement broad sense is on the basis of S conversion, introduce the Gaussian window width adjusting factor and the DC component place Gaussian window width adjusting factor and adopt the think of of fast algorithm conceivable.

Step S14: utilize the linear characteristic of improving the quick S conversion of broad sense, each first matrix of consequence is carried out to linearity and decompose, obtain second matrix of consequence corresponding to each frequency component of various discrete signal.

Step S15: each second matrix of consequence is carried out to the contrary quick S conversion of broad sense that improves according to inversion non-destructive respectively, obtain the reconstruct time-domain signal of each frequency component of various discrete signal.

Step S16: according to electric energy metrical requirement, each reconstruct time-domain signal is carried out to corresponding point multiply each other and result of product is cumulative, the result of product accumulated value of each reconstruct time-domain signal is added and obtains cumulative sum, described cumulative sum is multiplied by the A/D sampling interval time, obtain the power consumption value within the described A/D sampling interval time, to obtain the power consumption value in Preset Time.

In the present embodiment, Preset Time is divided by the A/D sampling interval time, and the value obtaining multiplies each other with the power consumption value within the described A/D sampling interval time, obtains the power consumption value in Preset Time.

In this application, introduced the Gaussian window width adjusting factor and the DC component place Gaussian window width adjusting factor owing to improving the quick S conversion of broad sense on the basis of S conversion, therefore Gaussian window shape can be adjusted according to actual conditions.

Convert based on fast algorithm owing to improving the quick S of broad sense, therefore in the time adopting improvement generalized S-transform, improved arithmetic speed, and in conjunction with utilizing the linear characteristic of improving the quick S conversion of broad sense to carry out linearity decomposition, reduced calculated amount, improved operation efficiency.

Embodiment bis-

Another embodiment

In the present embodiment, what illustrate is respectively various discrete signal to be improved to the quick S conversion of broad sense, obtains the process of each self-corresponding the first matrix of consequence of various discrete signal.

Owing to various discrete signal being improved to the quick S conversion of broad sense, the process that obtains each self-corresponding the first matrix of consequence of various discrete signal is identical, therefore in the present embodiment, only any one discrete signal is improved to the quick S conversion of broad sense, the process that obtains the first matrix of consequence that this discrete signal is corresponding is described.

In the present embodiment, described discrete signal is improved to the quick S conversion of broad sense, the detailed process that obtains the first matrix of consequence that this discrete signal is corresponding is:

Adopt the quick S conversion of discrete improvement broad sense expression formula $\left\{\begin{array}{c}S[a,n]=\underset{p=-{\mathrm{width}}_n\left(n\right)}{\overset{{\mathrm{width}}_n\left(n\right)-1}{\mathrm{\Σ}}}H[p+{\mathrm{centre}}_n\left(n\right)]e^{-\frac{{2\mathrm{\π}}^2{(p+\frac{\mathrm{N\γ}}{f_s})}^{2\mathrm{\β}}{\mathrm{\α}}^2}{{{\mathrm{centre}}_v}^2\left(n\right)}}{e}^{j\frac{2\mathrm{\πpa}}{N}}\\ S[a,0]=\frac{1}{N}\underset{K=0}{\overset{N-1}{\mathrm{\Σ}}}h\left(\mathrm{k\Δt}\right)\end{array}\right.$ Described discrete signal is improved to the quick S conversion of broad sense, obtain the first matrix of consequence, described in $\begin{array}{c}S[a,n]=\underset{p=-{\mathrm{width}}_n\left(n\right)}{\overset{{\mathrm{width}}_n\left(n\right)-1}{\mathrm{\Σ}}}H[p+{\mathrm{centre}}_n\left(n\right)]e^{-\frac{{2\mathrm{\π}}^2{(p+\frac{\mathrm{N\γ}}{f_s})}^{2\mathrm{\β}}{\mathrm{\α}}^2}{{{\mathrm{centre}}_v}^2\left(n\right)}}{e}^{j\frac{2\mathrm{\πpa}}{N}}\end{array}$ For the quick S conversion of the discrete improvement broad sense expression formula of AC portion, described in for the quick S conversion of the discrete improvement broad sense expression formula of direct current component, a=0,1 ... N-1, n=1 ... n _max-1, p=-width (n) ..., 0 ... width (n)-1, the sampling number that described N is electric signal to be analyzed, the sampling time interval that described Δ t is electric signal to be analyzed, (k Δ is t) sampled value of each sampling instant to described h.

In the present embodiment, by $\begin{array}{c}S[a,n]=\underset{p=-{\mathrm{width}}_n\left(n\right)}{\overset{{\mathrm{width}}_n\left(n\right)-1}{\mathrm{\Σ}}}H[p+{\mathrm{centre}}_n\left(n\right)]e^{-\frac{{2\mathrm{\π}}^2{(p+\frac{\mathrm{N\γ}}{f_s})}^{2\mathrm{\β}}{\mathrm{\α}}^2}{{{\mathrm{centre}}_v}^2\left(n\right)}}{e}^{j\frac{2\mathrm{\πpa}}{N}}\end{array}$ The AC compounent of electric signal is carried out to discrete improvement broad sense Fast transforms, obtain the first matrix of consequence of AC compounent.

By the DC component of electric signal is carried out to discrete improvement broad sense Fast transforms, obtain the first matrix of consequence of DC component.

Wherein, the quick S conversion of described discrete improvement broad sense expression formula obtains by the quick S transformation for mula of improvement broad sense is carried out to discrete transform, and the quick S transformation for mula of described improvement broad sense is $S(t,f)={\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)w_m(t-\mathrm{\τ},\mathrm{\α},\mathrm{\β},\mathrm{\γ})e^{-j2\mathrm{\πf\τ}}\mathrm{d\τ},$ Described $w_m(t-\mathrm{\τ},\mathrm{\α},\mathrm{\β},\mathrm{\γ})=\frac{\mathrm{\α}{\left(\right|f|+\mathrm{\γ})}^{\mathrm{\β}}}{\sqrt{2\mathrm{\π}}}{e}^{-\frac{{\mathrm{\α}}^2{\left(\right|f|+\mathrm{\γ})}^{2\mathrm{\β}}{(t-\mathrm{\τ})}^2}{2}},$ H (τ) is electric signal to be analyzed, and t is time shift variable, and f is frequency, and τ is the time, and α and β are the Gaussian window width adjusting factors, and γ is the DC component place Gaussian window width adjusting factor, w _m(t-τ, α, beta, gamma) is to improve Generalized Gaussian window.

Improve the quick S transformation for mula of broad sense than S transformation for mula of the prior art $S(t,f)={\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)w(t-\mathrm{\τ},\mathrm{\σ})e^{-j2\mathrm{\πf\τ}}\mathrm{d\τ}$ α, β and γ are increased.

The process of in the present embodiment, carrying out discrete transform to improving broad sense quick S transformation for mula is:

Steps A: utilize formula calculate the frequency domain radius of described improvement Generalized Gaussian window, described in for described frequency domain radius.

Step B: the frequency values of calculated rate sampled point number and center frequency points, described frequency sampling is counted out and is designated as n _max, the frequency values of described center frequency points is designated as f _n, described center frequency points is frequency sampling point.

In the present embodiment, step B comprises:

B11: the frequency that makes n frequency sampling point is f _n(n is since 0), the relation of adjacent two centre frequencies: $f_n+\frac{\mathrm{\α}{\left(\right|f_n|+\mathrm{\γ})}^{\mathrm{\β}}}{2\sqrt{2}\mathrm{\π}}+\frac{\mathrm{\α}{\left(\right|f_{n+1}|+\mathrm{\γ})}^{\mathrm{\β}}}{2\sqrt{2}\mathrm{\π}}=f_{n+1}.$

B12: making first frequency central point is that dc point is f ₀=0, and foundation obtain the frequency values f of each center frequency points _n, n=0 ..., n _max-1.

B13: solve inequality obtain the frequency sampling n that counts out _max, described f _sfor the sample frequency of described electric signal to be analyzed.

Step C: the discrete radius of improvement Generalized Gaussian window that calculates each center frequency points is described is width _n(n) improve the discrete radius of Generalized Gaussian window, to round symbol.

Step D: the shift length that calculates the frequency spectrum of each center frequency points is described centre _nfor shift length.

Step e: calculate the coverage of each frequency domain of described improvement Generalized Gaussian window, coverage is by f _startand f (n) _end(n) characterize, wherein, f _start(0)=0, f _start(n+1)=centre _n(n)+width _n(n)+width _n(n+1), n=0 ..., n _max-1, f _end(n)=f _start(n)+2width _n(n).

Step F: the FFT (fast Fourier transform of calculating described electric signal to be analyzed, Fast Fourier Transformation) result, and described FFT result is shifted, obtain a FFT result, be designated as H[p], direct current result in a described FFT result intermediate frequency spectrum information is positioned at center, and negative frequency component and positive frequency component lay respectively at both sides.

Step G: the time domain radius that calculates the improvement Generalized Gaussian window of each center frequency points is

$W_m[p,\mathrm{centre}\left(n\right)]=\exp [-\frac{2{\mathrm{\π}}^2{(p+\frac{\mathrm{N\γ}}{f_s})}^{2\mathrm{\β}}{\mathrm{\α}}^2}{{{\mathrm{centre}}_n}^2\left(n\right)}],p=-{\mathrm{width}}_n\left(n\right),-{\mathrm{width}}_n\left(n\right)+1,...,0,...,{\mathrm{width}}_n\left(n\right)-1.$

Step H: a described FFT result is added with the shift length of the frequency spectrum of each center frequency points respectively, obtains each self-corresponding the 2nd FFT result of each center frequency points, be designated as H[p+centre _n(n)].

Step I: the 2nd FFT result separately of each center frequency points and the time domain radius of its improvement Generalized Gaussian window are separately multiplied each other, obtain the operation result of each center frequency points, each operation result is arranged according to distributing position in FFT frequency spectrum, obtain rank results, described rank results is carried out to IFFT operation, obtain the quick S conversion of the discrete improvement broad sense expression formula of AC portion.

Embodiment tri-

In the present embodiment, what illustrate is to utilize the linear characteristic of improving the quick S of broad sense conversion, and each first matrix of consequence is carried out to linearity decomposition, obtains the process of the second matrix of consequence corresponding to each frequency component of various discrete signal.

Decompose owing to each first matrix of consequence being carried out to linearity, the process that obtains the second matrix of consequence corresponding to each frequency component of various discrete signal is identical, therefore in the present embodiment, only any one first matrix of consequence is carried out to linearity and decompose, the process that obtains the second matrix of consequence corresponding to each frequency component of the discrete signal under this first matrix of consequence is described.

In the present embodiment, described the first matrix of consequence is carried out to linearity and decompose, the process that obtains the second matrix of consequence corresponding to each frequency component of the discrete signal under this first matrix of consequence is specially:

Utilize the linear characteristic S[h (t) that improves the quick S conversion of broad sense]=S[x (t)+y (t)]=S[x (t)]+S[y (t)], in the time calculating certain frequency component, only retain in described the first matrix of consequence part that should frequency component, making part assignment corresponding to the frequency component except this frequency component in described the first matrix of consequence is 0.

Described h (t), x (t), y (t) are signal to be analyzed and h (t)=x (t)+y (t).

Now illustrate linear process of decomposing, as the first matrix of consequence S[a of signal h to be analyzed (t), n], a is the time, n is frequency, when calculating the n of signal to be analyzed ₀when frequency component, only need to retain the S[a of the first matrix of consequence, n ₀] part, except n ₀part assignment corresponding to all the other frequency components outside frequency component is 0, the n obtaining ₀the second matrix of consequence of frequency component is:

Embodiment tetra-

In the present embodiment, what illustrate is that each second matrix of consequence is carried out to the contrary quick S conversion of broad sense that improves according to inversion non-destructive respectively, obtains the process of the reconstruct time-domain signal of each frequency component of various discrete signal.

Owing to converting each second matrix of consequence is carried out to the contrary quick S of broad sense that improves according to inversion non-destructive respectively, the process of reconstruct time-domain signal that obtains each frequency component of various discrete signal is identical, therefore in the present embodiment, only any one second matrix of consequence is carried out to the contrary quick S conversion of broad sense that improves according to inversion non-destructive, the process that obtains the reconstruct time-domain signal of each frequency component of the affiliated discrete signal of this second matrix of consequence is described.

Any one second matrix of consequence is carried out to the contrary quick S conversion of broad sense that improves according to inversion non-destructive, and the inversion non-destructive that obtains the reconstruct time-domain signal institute foundation of each frequency component of the affiliated discrete signal of this second matrix of consequence is: ${\∫}_{-\∞}^{+\∞}{w}_m(t-\mathrm{\τ})\mathrm{dt}={\∫}_{-\∞}^{+\∞}\frac{1}{\mathrm{\σ}\sqrt{2\mathrm{\π}}}{e}^{-\frac{{(t-\mathrm{\τ})}^2}{2{\mathrm{\σ}}^2}}\mathrm{dt}=1,$ $\begin{array}{c}{\∫}_{-\∞}^{+\∞}S(t,f)\mathrm{dt}={\∫}_{-\∞}^{+\∞}\left[{\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)w_m(t-\mathrm{\τ})e^{-j2\mathrm{\πf\τ}}\mathrm{d\τ}\right]\mathrm{dt}={\∫}_{-\∞}^{+\∞}\left[{\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)w_m(t-\mathrm{\τ})e^{-j2\mathrm{\πf\τ}}\mathrm{dt}\right]\mathrm{d\τ}\\ ={\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)e^{-j2\mathrm{\πf\τ}}\left[{\∫}_{-\∞}^{+\∞}{w}_m(t-\mathrm{\τ})\mathrm{dt}\right]\mathrm{d\τ}={\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)e^{-j2\mathrm{\πf\τ}}\mathrm{d\τ}=H\left(f\right)\end{array},$ $h\left(\mathrm{\τ}\right)={\∫}_{-\∞}^{+\∞}H\left(f\right)e^{j2\mathrm{\πf\τ}}\mathrm{d\τ}={\∫}_{-\∞}^{+\∞}\left[{\∫}_{-\∞}^{+\∞}S(t,f)\mathrm{dt}\right]e^{j2\mathrm{\πf\τ}}\mathrm{d\τ}.$

Any one second matrix of consequence is carried out to the contrary quick S conversion of broad sense that improves according to inversion non-destructive, and the contrary principle of improving the quick S conversion of broad sense that obtains the reconstruct time-domain signal institute foundation of each frequency component of the affiliated discrete signal of this second matrix of consequence is:

$\begin{array}{c}\mathrm{\α}(v,f)={\∫}_{-\∞}^{+\∞}S(t,f)e^{-j2\mathrm{\πvt}}\mathrm{dt}=\\ {\∫}_{-\∞}^{+\∞}\left[{\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)w_m(t-\mathrm{\τ})e^{-j2\mathrm{\πf\τ}}\mathrm{d\τ}\right]e^{-j2\mathrm{\πvt}}\mathrm{dt}=\\ {\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)\left[{\∫}_{-\∞}^{+\∞}{w}_m(t-\mathrm{\τ})e^{-j2\mathrm{\πvt}}\mathrm{dt}\right]e^{-j2\mathrm{\πf\τ}}\mathrm{d\τ}=\\ {\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)\left[{\∫}_{-\∞}^{+\∞}{w}_m\left(t\right)e^{-j2\mathrm{\πv}(t+\mathrm{\τ})}\mathrm{dt}\right]e^{-j2\mathrm{\πf\τ}}\mathrm{d\τ}=\\ {\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)\left[{\∫}_{-\∞}^{+\∞}{w}_m\left(t\right)e^{-j2\mathrm{\πvt}}\mathrm{dt}\right]e^{-j2\mathrm{\πv\τ}}{e}^{-j2\mathrm{\πf\τ}}\mathrm{d\τ}=\\ {\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)W_m\left(v\right)e^{-j2\mathrm{\πv\τ}}{e}^{-j2\mathrm{\πf\τ}}\mathrm{d\τ}=H(v+f)W_m\left(v\right)\end{array},$

H(v+f)＝α(v，f)/W _m(v)

Described a (v, f) be the Fourier transform result to described time shift variable t and the Fourier transform result to described time τ, described v is the Fourier transform to described time shift variable t, described f is the Fourier transform to described τ, and H (v+f) is the Fourier transform result to electric signal to be analyzed.

In the present embodiment, H (v+f) displacement is obtained to H (v), then H (v) is carried out to IFFT operation, obtain h (τ).

In the present embodiment, the detailed process of the time-domain signal of the some discrete signals of reconstruct can comprise the steps:

1) setting up a length is array $\mathrm{gauss}[1+\frac{N}{2}]\left(\mathrm{gauss}\right[0]=1)$ With wherein array the quick S transformation results of improvement broad sense (numbering of array is since 0) of depositing DC component and positive frequency component, its corresponding frequency range is so wherein be numbered the element of i (i.e. i element) corresponding to frequency in addition set up the array H[N that length is N] to deposit the FFT result of signal;

2)For?n＝0，2，3，…，n _max-1：

To pos_spe[f _start(n)+1] to pos_spe[f _end(n)] be total to 2width _n(n) individual element carries out FFT operation, and is multiplied by coefficient and still leave result in pos_spe[f _start(n)+1] to pos_spe[f _end(n)];

This time-frequency domain improves Generalized Gaussian window sequence

p＝-width _n(n)，…，0，…，width _n(n)-1，

Place it in gauss[f _start(n)+1] to gauss[f _end(n)];

End?of?For。

3) by array middle element divided by the element of middle correspondence, and result is placed on in.

4) by array middle element is put into array H[N] middle corresponding position.

5)For $i=\mathrm{1,2},...,\frac{N}{2}-1$ ：

By element H[i] get its conjugation and be placed on H[N-i] in; End of For.

6) to H[N] carry out IFFT operation and be multiplied by coefficient N, finally obtain reconstruct time-domain signal.

It should be noted that, method provided by the invention is applicable to the electric energy metrical occasion under common stable state and unstable signal.For the specific implementation process of method provided by the invention is described, below the electric energy metrical under harmonic wave, a harmonic signal is analyzed as emphasis.

Make signal sampling frequency 6.4kHz, emulation duration 0.2s.The voltage, the current signal that carry out electric energy metrical comprise harmonic wave and a harmonic wave.While carrying out electric energy metrical so, comprise following some: first voltage, current signal are carried out to low-pass filtering treatment, obtain simulating signal, for A/D sampling is prepared; Then, simulating signal is carried out to A/D sampling and obtain discrete signal, for next step signal processing lays the foundation; Afterwards, adopt the quick S transfer pair of improvement broad sense voltage, electric current to process respectively the matrix of consequence that obtains voltage, electric current, by linear characteristic, matrix of consequence is separated afterwards, thereby represented respectively the matrix of consequence of first-harmonic composition and other compositions, then the matrix of consequence after separating is carried out to the contrary quick S of broad sense that improves and convert the fundamental signal and the distorted signal that obtain respectively voltage, electric current; Finally according to formula solve all kinds of energy values.Metering result is as shown in table 1, table 2, and table 1 is the electric energy metrical analysis of simulation result of harmonic wave, a harmonic signal, and table 2 is traditional S conversion and comparison operation time of S conversion fast.

Wherein voltage signal is:

$\begin{array}{c}u\left(t\right)=\sqrt{2}\×220\×\sin \left(2\mathrm{\π}{f}_{0}t\right)+\sqrt{2}\×11\×\sin (2\mathrm{\π}\×3f_{0}t)+\\ \sqrt{2}\×\sin (2\mathrm{\π}\×3.5f_{0}t)+\sqrt{2}\×5\×\sin (2\mathrm{\π}\×5f_{0}t)\end{array}.$

Current signal is:

Table 1

Table 2

Table 1, table 2 have shown can realize accurate-metering to electric energy and measuring accuracy higher than wavelet package transforms based on the electric energy gauging method that improves the quick S of broad sense conversion, and compared to S conversion, the inventive method has greatly shortened operation time simultaneously.

Embodiment five

In the present embodiment, the electric power meter based on improving the quick S conversion of broad sense that provides the application to provide, refer to Fig. 2, a kind of structural representation of the electric power meter that the quick S of improvement broad sense that it shows the application provides converts, the electric power meter based on improving the quick S conversion of broad sense comprises: filtration module 21, sampling module 22, the first conversion module 23, decomposing module 24, the second conversion module 25 and computing module 26.

Filtration module 21, for the electrical network electric signal that need to carry out electric energy metrical is carried out to low-pass filtering, obtains simulating signal.

Sampling module 22, for described simulating signal is carried out to A/D sampling, obtains multiple discrete signals.

The first conversion module 23, for respectively various discrete signal being improved to the quick S conversion of broad sense, obtain each self-corresponding the first matrix of consequence of various discrete signal, the quick S conversion of described improvement broad sense is on the basis of S conversion, introduce the Gaussian window width adjusting factor and the DC component place Gaussian window width adjusting factor and adopt the think of of fast algorithm conceivable.

In the present embodiment, the first conversion module 23 comprises: the first converter unit, and for adopting the quick S conversion of discrete improvement broad sense expression formula $\left\{\begin{array}{c}S[a,n]=\underset{p={-\mathrm{width}}_n\left(n\right)}{\overset{{\mathrm{width}}_n\left(n\right)-1}{\mathrm{\Σ}}}H[p+{\mathrm{centre}}_n\left(n\right)]e^{-\frac{{2\mathrm{\π}}^2{(p+\frac{\mathrm{N\γ}}{f_s})}^{2\mathrm{\β}}{\mathrm{\α}}^2}{{{\mathrm{centre}}_v}^2\left(n\right)}}{e}^{j\frac{2\mathrm{\πpa}}{N}}\\ S[a,0]=\frac{1}{N}\underset{K=0}{\overset{N-1}{\mathrm{\Σ}}}h\left(\mathrm{k\Δt}\right)\end{array}\right.$

Various discrete signal is improved to the quick S conversion of broad sense, obtain each self-corresponding the first matrix of consequence of various discrete signal, described in $S[a,n]=\underset{p={-\mathrm{width}}_n\left(n\right)}{\overset{{\mathrm{width}}_n\left(n\right)-1}{\mathrm{\Σ}}}H[p+{\mathrm{centre}}_n\left(n\right)]e^{-\frac{{2\mathrm{\π}}^2{(p+\frac{\mathrm{N\γ}}{f_s})}^{2\mathrm{\β}}{\mathrm{\α}}^2}{{{\mathrm{centre}}_v}^2\left(n\right)}}{e}^{j\frac{2\mathrm{\πpa}}{N}}$ For the quick S conversion of the discrete improvement broad sense expression formula of AC portion, a=0,1 ... N-1, n=1 ... n _max-1, p=-width (n) ..., 0 ... width (n)-1, described in for the quick S conversion of the discrete improvement broad sense expression formula of direct current component, the sampling number that described N is electric signal to be analyzed, the sampling time interval that described Δ t is electric signal to be analyzed, (k Δ is t) sampled value of each sampling instant to described h;

The quick S conversion of described discrete improvement broad sense expression formula obtains by the quick S transformation for mula of improvement broad sense is carried out to discrete transform, and the quick S transformation for mula of described improvement broad sense is $S(t,f)={\∫}_{-\∞}^{+\∞}h\left(\mathrm{\τ}\right)w_m(t-\mathrm{\τ},\mathrm{\α},\mathrm{\β},\mathrm{\γ})e^{-j2\mathrm{\πf\τ}}\mathrm{d\τ},$ Described $w_m(t-\mathrm{\τ},\mathrm{\α},\mathrm{\β},\mathrm{\γ})=\frac{\mathrm{\α}{\left(\right|f|+\mathrm{\γ})}^{\mathrm{\β}}}{\sqrt{2\mathrm{\π}}}{e}^{-\frac{{\mathrm{\α}}^2{\left(\right|f|+\mathrm{\γ})}^{2\mathrm{\β}}{(t-\mathrm{\τ})}^2}{2}},$ H (τ) is electric signal to be analyzed, and t is time shift variable, and f is frequency, and τ is the time, and α and β are the Gaussian window width adjusting factors, and γ is the DC component place Gaussian window width adjusting factor, w _m(t-τ, α, beta, gamma) is to improve Generalized Gaussian window.

Decomposing module 24, for utilizing the linear characteristic of improving the quick S conversion of broad sense, carries out linearity to each first matrix of consequence and decomposes, and obtains second matrix of consequence corresponding to each frequency component of various discrete signal.

The second conversion module 25, for each second matrix of consequence is carried out to the contrary quick S conversion of broad sense that improves according to inversion non-destructive respectively, obtains the reconstruct time-domain signal of each frequency component of various discrete signal.

Computing module 26, be used for according to electric energy metrical requirement, each reconstruct time-domain signal is carried out to corresponding point multiply each other and result of product is cumulative, the result of product accumulated value of each reconstruct time-domain signal is added and obtains cumulative sum, described cumulative sum is multiplied by the A/D sampling interval time, obtain the power consumption value within the described A/D sampling interval time, to obtain the power consumption value in Preset Time.

In the present embodiment, the first conversion module 23, decomposing module 24, the second conversion module 25 and computing module 26 can be integrated in same digital signal processor.

The electric power meter based on improving the quick S conversion of broad sense shown in Fig. 2 can also comprise: the 3rd conversion module, and for carrying out discrete transform to improving the quick S transformation for mula of broad sense.

Wherein, the 3rd conversion module comprises: the first computing unit, the second computing unit, the 3rd computing unit, the 4th computing unit, the 5th computing unit, the 6th computing unit, the 7th computing unit, the 8th computing unit and the 9th computing unit.

The first computing unit, for utilizing formula calculate the frequency domain radius of described improvement Generalized Gaussian window, described in for described frequency domain radius.

The second computing unit, for the frequency values of calculated rate sampled point number and center frequency points, described frequency sampling is counted out and is designated as n _max, the frequency values of described center frequency points is designated as f _n, described center frequency points is frequency sampling point.

In the present embodiment, the second computing unit specifically comprises: the first computation subunit, the second computation subunit and the 3rd computation subunit.

The first computation subunit, using the frequency of n frequency sampling point is up to the present f _n(n is since 0), the relation of adjacent two centre frequencies: $f_n+\frac{\mathrm{\α}{\left(\right|f_n|+\mathrm{\γ})}^{\mathrm{\β}}}{2\sqrt{2}\mathrm{\π}}+\frac{\mathrm{\α}{\left(\right|f_{n+1}|+\mathrm{\γ})}^{\mathrm{\β}}}{2\sqrt{2}\mathrm{\π}}=f_{n+1}.$

The second computation subunit, with first frequency central point be up to the present that dc point is f ₀=0, and foundation obtain the frequency values f of each center frequency points _n, n=0 ..., n _max-1.

The 3rd computation subunit, for solving inequality obtain the frequency sampling n that counts out _max, described f _sfor the sample frequency of described electric signal to be analyzed.

The 3rd computing unit, for the discrete radius of improvement Generalized Gaussian window that calculates each center frequency points is described is width _n(n) improve the discrete radius of Generalized Gaussian window, to round symbol.

The 4th computing unit, for the shift length of the frequency spectrum that calculates each center frequency points is described centre _nfor shift length.

The 5th computing unit, for calculating the coverage of each frequency domain of described improvement Generalized Gaussian window, coverage is by f _startand f (n) _end(n) characterize, wherein, f _start(0)=0, f _start(n+1)=centre _n(n)+width _n(n)+width _n(n+1), n=0 ..., n _max-1, f _end(n)=f _start(n)+2width _n(n).

The 6th computing unit, for calculating the FFT result of described electric signal to be analyzed, and is shifted to described FFT result, obtain a FFT result, be designated as H[p], the direct current result in a described FFT result intermediate frequency spectrum information is positioned at center, and negative frequency component and positive frequency component lay respectively at both sides.

The 7th computing unit, for the time domain radius of the improvement Generalized Gaussian window that calculates each center frequency points is

$W_m[p,\mathrm{centre}\left(n\right)]=\exp [-\frac{2{\mathrm{\π}}^2{(p+\frac{\mathrm{N\γ}}{f_s})}^{2\mathrm{\β}}{\mathrm{\α}}^2}{{{\mathrm{centre}}_n}^2\left(n\right)}],p=-{\mathrm{width}}_n\left(n\right),-{\mathrm{width}}_n\left(n\right)+1,...,0,...,{\mathrm{width}}_n\left(n\right)-1$

The 8th computing unit, for a described FFT result is added with the shift length of the frequency spectrum of each center frequency points respectively, obtains each self-corresponding the 2nd FFT result of each center frequency points, is designated as H[p+centre _n(n)].

The 9th computing unit, for the 2nd FFT result separately of each center frequency points and the time domain radius of its improvement Generalized Gaussian window are separately multiplied each other, obtain the operation result of each center frequency points, each operation result is arranged according to distributing position in FFT frequency spectrum, obtain rank results, described rank results is carried out to IFFT operation, obtain the quick S conversion of the discrete improvement broad sense expression formula of AC portion.

In the present embodiment, described decomposing module 24 is specifically for utilizing the linear characteristic S[h (t) that improves the quick S conversion of broad sense]=S[x (t)+y (t)]=S[x (t)]+S[y (t)], in the time calculating certain frequency component, only retain in described the first matrix of consequence part that should frequency component, making part assignment corresponding to the frequency component except this frequency component in described the first matrix of consequence is 0;

Described h (t), x (t), y (t) are signal to be analyzed and h (t)=x (t)+y (t)

In the present embodiment, computing module 26 specifically comprises: separative element and the tenth computing unit.

Separative element, for this reconstruct time-domain signal is separated, obtains each harmonic voltage, electric current discrete signal, is designated as respectively u _n[k], i _n[k].

The tenth computing unit, for according to formula calculate the electric energy of this reconstruct time-domain signal unit interval internal consumption.

It should be noted that, each embodiment in this instructions all adopts the mode of going forward one by one to describe, and what each embodiment stressed is and the difference of other embodiment, between each embodiment identical similar part mutually referring to.For device class embodiment, because it is substantially similar to embodiment of the method, so description is fairly simple, relevant part is referring to the part explanation of embodiment of the method.

Finally, also it should be noted that, in this article, relational terms such as the first and second grades is only used for an entity or operation to separate with another entity or operational zone, and not necessarily requires or imply and between these entities or operation, have the relation of any this reality or sequentially.And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thereby the process, method, article or the equipment that make to comprise a series of key elements not only comprise those key elements, but also comprise other key elements of clearly not listing, or be also included as the intrinsic key element of this process, method, article or equipment.The in the situation that of more restrictions not, the key element being limited by statement " comprising ... ", and be not precluded within process, method, article or the equipment that comprises described key element and also have other identical element.

The electric energy gauging method and the device that above the application are provided are described in detail, applied principle and the embodiment of specific case to the application herein and set forth, the explanation of above embodiment is just for helping to understand the application's method and core concept thereof; , for one of ordinary skill in the art, according to the application's thought, all will change in specific embodiments and applications, in sum, this description should not be construed as the restriction to the application meanwhile.

Claims (13)

1. an electric energy gauging method, is characterized in that, comprising:

Carry out low-pass filtering to carrying out electric signal in the electrical network of electric energy metrical, obtain simulating signal;

Described simulating signal is carried out to A/D sampling, obtain multiple discrete signals;

Respectively various discrete signal is improved to the quick S conversion of broad sense, obtain each self-corresponding the first matrix of consequence of various discrete signal, the quick S conversion of described improvement broad sense is on the basis of S conversion, introduce the Gaussian window width adjusting factor and the DC component place Gaussian window width adjusting factor and adopt the think of of fast algorithm conceivable;

Utilize the linear characteristic of improving the quick S conversion of broad sense, each first matrix of consequence is carried out to linearity and decompose, obtain second matrix of consequence corresponding to each frequency component of various discrete signal;

Each second matrix of consequence is carried out to the contrary quick S conversion of broad sense that improves according to inversion non-destructive respectively, obtain the reconstruct time-domain signal of each frequency component of various discrete signal;

According to electric energy metrical requirement, each reconstruct time-domain signal is carried out to corresponding point multiply each other and result of product is cumulative, the result of product accumulated value of each reconstruct time-domain signal is added and obtains cumulative sum, described cumulative sum is multiplied by the A/D sampling interval time, obtain the power consumption value within the described A/D sampling interval time, to obtain the power consumption value in Preset Time.

2. method according to claim 1, is characterized in that, respectively various discrete signal is improved to the quick S conversion of broad sense, obtains the process of each self-corresponding the first matrix of consequence of various discrete signal, comprising:

Adopt the quick S conversion of discrete improvement broad sense expression formula various discrete signal is improved to the quick S conversion of broad sense, obtain each self-corresponding the first matrix of consequence of various discrete signal, described in for the quick S conversion of the discrete improvement broad sense expression formula of AC portion, described in for the quick S conversion of the discrete improvement broad sense expression formula of direct current component, a=0,1 ... N-1, n=1 ... n _max-1, p=-width (n) ..., 0 ... width (n)-1, the sampling number that described N is electric signal to be analyzed, the sampling time interval that described Δ t is electric signal to be analyzed, (k Δ is t) sampled value of each sampling instant to described h.

3. method according to claim 2, is characterized in that, the quick S conversion of described discrete improvement broad sense expression formula obtains by the quick S transformation for mula of improvement broad sense is carried out to discrete transform, and the quick S transformation for mula of described improvement broad sense is described h (τ) is electric signal to be analyzed, and t is time shift variable, and f is frequency, and τ is the time, and α and β are the Gaussian window width adjusting factors, and γ is the DC component place Gaussian window width adjusting factor, w _m(t-τ, α, beta, gamma) is to improve Generalized Gaussian window;

The process of wherein, carrying out discrete transform to improving broad sense quick S transformation for mula is:

A: utilize formula calculate the frequency domain radius of described improvement Generalized Gaussian window, described in for described frequency domain radius;

B: the frequency values of calculated rate sampled point number and center frequency points, described frequency sampling is counted out and is designated as n _max, the frequency values of described center frequency points is designated as f _n, described center frequency points is frequency sampling point;

C: the discrete radius of improvement Generalized Gaussian window that calculates each center frequency points is described is width _n(n) improve the discrete radius of Generalized Gaussian window, to round symbol;

D: the shift length that calculates the frequency spectrum of each center frequency points is described centre _nfor shift length;

E: calculate the coverage of each frequency domain of described improvement Generalized Gaussian window, coverage is by f _startand f (n) _end(n) characterize, wherein, f _start(0)=0, f _start(n+1)=centre _n(n)+width _n(n)+width _n(n+1), n=0 ..., n _max-1, f _end(n)=f _start(n)+2width _n(n);

F: the fast Fourier transform FFT result of calculating described electric signal to be analyzed, and described FFT result is shifted, obtain a FFT result, be designated as H[p], direct current result in a described FFT result intermediate frequency spectrum information is positioned at center, and negative frequency component and positive frequency component lay respectively at both sides;

G: the time domain radius that calculates the improvement Generalized Gaussian window of each center frequency points is p=-width _n(n) ,-width _n(n)+1 ..., 0 ..., width _n(n)-1;

H: a described FFT result is added with the shift length of the frequency spectrum of each center frequency points respectively, obtains each self-corresponding the 2nd FFT result of each center frequency points, be designated as H[p+centre _n(n)];

I: the 2nd FFT result separately of each center frequency points and the time domain radius of its improvement Generalized Gaussian window are separately multiplied each other, obtain the operation result of each center frequency points, each operation result is arranged according to distributing position in FFT frequency spectrum, obtain rank results, described rank results is carried out to IFFT operation, obtain the quick S conversion of the discrete improvement broad sense expression formula of AC portion.

4. method according to claim 3, is characterized in that, described step B comprises:

B11: the frequency that makes n frequency sampling point is f _n(n is since 0), the relation of adjacent two centre frequencies:

B12: making first frequency central point is that dc point is f ₀=0, and foundation obtain the frequency values f of each center frequency points _n, n=0 ..., n _max-1;

B13: solve inequality obtain the frequency sampling n that counts out _max, described f _sfor the sample frequency of described electric signal to be analyzed.

5. method according to claim 3, it is characterized in that, utilize the linear characteristic of improving the quick S conversion of broad sense, each first matrix of consequence is carried out to linearity and decompose, the process that obtains the second matrix of consequence corresponding to each frequency component of various discrete signal, comprising:

Utilize the linear characteristic S[h (t) that improves the quick S conversion of broad sense]=S[x (t)+y (t)]=S[x (t)]+S[y (t)], in the time calculating certain frequency component of each the first matrix of consequence, only retain in described the first matrix of consequence part that should frequency component, making part assignment corresponding to the frequency component except this frequency component in described the first matrix of consequence is 0;

Described h (t), x (t), y (t) are signal to be analyzed and h (t)=x (t)+y (t).

6. method according to claim 3, is characterized in that, described inversion non-destructive is:

The described contrary principle of improving the quick S conversion of broad sense is

H(v+f)＝α(v，f)/W _m(v)，

Described a (v, f) be the Fourier transform result to described time shift variable t and the Fourier transform result to described time τ, described v is the Fourier transform to described time shift variable t, described f is the Fourier transform to described τ, and H (v+f) is the Fourier transform result to electric signal to be analyzed.

7. method according to claim 6, is characterized in that, according to electric energy metrical requirement, any one reconstruct time-domain signal is carried out that corresponding point multiply each other and by process cumulative result of product, being comprised:

This reconstruct time-domain signal is separated, obtain each harmonic voltage, electric current discrete signal, be designated as respectively u _n[k], i _n[k];

According to formula calculate the electric energy of this reconstruct time-domain signal unit interval internal consumption.

8. an electric power meter, is characterized in that, comprising:

Filtration module, for the electrical network electric signal that need to carry out electric energy metrical is carried out to low-pass filtering, obtains simulating signal;

Sampling module, for described simulating signal is carried out to A/D sampling, obtains multiple discrete signals;

The first conversion module, for respectively various discrete signal being improved to the quick S conversion of broad sense, obtain each self-corresponding the first matrix of consequence of various discrete signal, the quick S conversion of described improvement broad sense is on the basis of S conversion, introduce the Gaussian window width adjusting factor and the DC component place Gaussian window width adjusting factor and adopt the think of of fast algorithm conceivable;

Decomposing module, for utilizing the linear characteristic of improving the quick S conversion of broad sense, carries out linearity to each first matrix of consequence and decomposes, and obtains second matrix of consequence corresponding to each frequency component of various discrete signal;

The second conversion module, for each second matrix of consequence is carried out to the contrary quick S conversion of broad sense that improves according to inversion non-destructive respectively, obtains the reconstruct time-domain signal of each frequency component of various discrete signal;

Computing module, be used for according to electric energy metrical requirement, each reconstruct time-domain signal is carried out to corresponding point multiply each other and result of product is cumulative, the result of product accumulated value of each reconstruct time-domain signal is added and obtains cumulative sum, described cumulative sum is multiplied by the A/D sampling interval time, obtain the power consumption value within the described A/D sampling interval time, to obtain the power consumption value in Preset Time.

9. device according to claim 8, is characterized in that, described the first conversion module comprises:

The first converter unit, for adopting the quick S conversion of discrete improvement broad sense expression formula

Various discrete signal is improved to the quick S conversion of broad sense, obtain each self-corresponding the first matrix of consequence of various discrete signal, described in for the quick S conversion of the discrete improvement broad sense expression formula of AC portion, described in for the quick S conversion of the discrete improvement broad sense expression formula of direct current component, a=0,1 ... N-1, n=1 ... n _max-1, p=-width (n) ..., 0 ... width (n)-1, the sampling number that described N is electric signal to be analyzed, the sampling time interval that described Δ t is electric signal to be analyzed, (k Δ is t) sampled value of each sampling instant to described h;

The quick S conversion of described discrete improvement broad sense expression formula obtains by the quick S transformation for mula of improvement broad sense is carried out to discrete transform, and the quick S transformation for mula of described improvement broad sense is described h (τ) is electric signal to be analyzed, and t is time shift variable, and f is frequency, and τ is the time, and α and β are the Gaussian window width adjusting factors, and γ is the DC component place Gaussian window width adjusting factor, w _m(t-τ, α, beta, gamma) is to improve Generalized Gaussian window.

10. device according to claim 9, is characterized in that, also comprises:

The 3rd conversion module, for carrying out discrete transform to improving the quick S transformation for mula of broad sense;

Described the 3rd conversion module comprises:

The first computing unit, for utilizing formula calculate the frequency domain radius of described improvement Generalized Gaussian window, described in for described frequency domain radius;

The second computing unit, for the frequency values of calculated rate sampled point number and center frequency points, described frequency sampling is counted out and is designated as n _max, the frequency values of described center frequency points is designated as f _n, described center frequency points is frequency sampling point;

The 3rd computing unit, for the discrete radius of improvement Generalized Gaussian window that calculates each center frequency points is described is width _n(n) improve the discrete radius of Generalized Gaussian window, to round symbol;

The 4th computing unit, for the shift length of the frequency spectrum that calculates each center frequency points is described centre _nfor shift length;

The 5th computing unit, for calculating the coverage of each frequency domain of described improvement Generalized Gaussian window, coverage is by f _startand f (n) _end(n) characterize, wherein, f _start(0)=0, f _start(n+1)=centre _n(n)+width _n(n)+width _n(n+1), n=0 ..., n _max-1, f _end(n)=f _start(n)+2width _n(n);

The 6th computing unit, for calculating the FFT result of described electric signal to be analyzed, and is shifted to described FFT result, obtain a FFT result, be designated as H[p], the direct current result in a described FFT result intermediate frequency spectrum information is positioned at center, and negative frequency component and positive frequency component lay respectively at both sides;

The 7th computing unit, for the time domain radius of the improvement Generalized Gaussian window that calculates each center frequency points is p=-width _n(n) ,-width _n(n)+1 ..., 0 ..., width _n(n)-1;

The 8th computing unit, for a described FFT result is added with the shift length of the frequency spectrum of each center frequency points respectively, obtains each self-corresponding the 2nd FFT result of each center frequency points, is designated as H[p+centre _n(n)];

The 9th computing unit, for the 2nd FFT result separately of each center frequency points and the time domain radius of its improvement Generalized Gaussian window are separately multiplied each other, obtain the operation result of each center frequency points, each operation result is arranged according to distributing position in FFT frequency spectrum, obtain rank results, described rank results is carried out to IFFT operation, obtain the quick S conversion of the discrete improvement broad sense expression formula of AC portion.

11. devices according to claim 10, is characterized in that, described the second computing unit comprises: the first computation subunit, using the frequency of n frequency sampling point is up to the present f _n(n is since 0), the relation of adjacent two centre frequencies:

The second computation subunit, with first frequency central point be up to the present that dc point is f ₀=0, and foundation obtain the frequency values f of each center frequency points _n, n=0 ..., n _max-1;

The 3rd computation subunit, for solving inequality obtain the frequency sampling n that counts out _max, described f _sfor the sample frequency of described electric signal to be analyzed.

12. devices according to claim 9, it is characterized in that, described decomposing module is specifically for utilizing the linear characteristic S[h (t) that improves the quick S conversion of broad sense]=S[x (t)+y (t)]=S[x (t)]+S[y (t)], in the time calculating certain frequency component, only retain in described the first matrix of consequence part that should frequency component, making part assignment corresponding to the frequency component except this frequency component in described the first matrix of consequence is 0;

Described h (t), x (t), y (t) are signal to be analyzed and h (t)=x (t)+y (t).

13. devices according to claim 12, is characterized in that, described computing module comprises:

Separative element, for this reconstruct time-domain signal is separated, obtains each harmonic voltage, electric current discrete signal, is designated as respectively u _n[k], i _n[k];

The tenth computing unit, for according to formula calculate the electric energy of this reconstruct time-domain signal unit interval internal consumption.