CN106021805A - Electric energy meter dynamic testing signal fitting method based on HHT analysis - Google Patents

Abstract

The invention provides an electric energy meter dynamic testing signal fitting method based on HHT analysis. The method comprises the steps that EMD decomposition is conducted on an actually-measured power grid waveform signal to obtain a modulating wave signal representing dynamic voltage fluctuation, a steady-state harmonic signal and an inter-harmonic signal separately; Hilbert transformation and inverse transformation are conducted on the modulating wave signal representing dynamic voltage fluctuation separately to obtain a new analysis signal, an instantaneous amplitude value signal of the analysis signal is solved, EMD decomposition is conducted on the instantaneous amplitude value to obtain amplitude-modulated wave signals, and amplitude value and frequency parameter information of dynamic voltage fluctuation is solved through Hilbert transformation; Hilbert transformation is directly conducted on the steady-state harmonic signal and the inter-harmonic signal to obtain amplitude value and frequency parameter information of harmonic waves and inter-harmonic waves; fitting is conducted through a least square method according to obtained waveform information of the dynamic voltage amplitude-modulated waves, the steady-state harmonic waves and the inter-harmonic waves to obtain a digital signal expression finally used for energy meter dynamic testing.

Description

A kind of electric energy meter dynamic measuring signal approximating method analyzed based on HHT

Technical field

The present invention relates to a kind of electric energy meter dynamic measuring signal approximating method analyzed based on HHT, belong to electric power instrument Table technical field of measurement and test.

Background technology

In recent years, China's intelligent grid has been enter into the all-round construction stage, and intelligent electric energy meter is as intelligent grid Important component part is widely used in power information acquisition system and applies.As connecting electrical network and visitor The key link at family, the electric energy metrical characteristic of intelligent electric energy meter, directly affect the economic profit of electric energy both sides of supply and demand Benefit.

Along with the fast development of intelligent grid, the metering performance of electric energy meter is proposed new requirement.One be by Introducing electrical network in new forms of energy such as wind energy, solar energy, tide energies, the novel power supply in electrical network presents relatively Larger Dynamic Characteristic.Two is that the astable load in power system gets more and more, and makes the power load in electrical network also present dynamic Step response.Due to being continuously increased of dynamic load, its caused electric energy metrical occurs that the problem of deviation becomes More and more prominent with important.

But, for a long time, the test to electric energy meter meter characteristic both at home and abroad is all at steady state test signal bar Carrying out under part, test signal is the most comprehensive with the reflection of Dynamic Signal composition to the stable state included in electrical network, Therefore, the matching of electric energy meter test signal needs to be more based on electrical network actual operating mode, just can make electric energy The analysis of table dynamic error characteristic is more accurately rationally.

Summary of the invention

It is an object of the invention to, the problem existed for existing electric energy meter test signal, the present invention proposes one The electric energy meter dynamic measuring signal approximating method analyzed based on HHT, it is possible to make test signal packet containing actual electric network Stable state and Dynamic Signal composition, more precisely reflect the practical operation situation of electrical network, dynamic to electric energy meter The analysis of error characteristics is more accurately rationally.

Realize technical scheme as follows:

A kind of electric energy meter dynamic measuring signal approximating method analyzed based on HHT, comprises the following steps:

(1) actual measurement network wave signal is carried out EMD decomposition, respectively obtain and characterize dynamic electric voltage fluctuation Modulating wave and stable state harmonic wave, m-Acetyl chlorophosphonazo signal, modulation wave signal frequency is fundamental frequency, stable state harmonic signal Frequency is first-harmonic integral multiple, and m-Acetyl chlorophosphonazo signal frequency is that first-harmonic is non-integral multiple；

(2) for characterizing the modulation wave signal that dynamic electric voltage fluctuates, it is carried out respectively Hilbert conversion with anti- Conversion, obtains new analytic signal, solves its instantaneous amplitude, then instantaneous amplitude signal is carried out EMD decomposition, Obtain each amplitude-modulated wave signal, solved amplitude and the frequency parameter letter of dynamic electric voltage fluctuation by Hilbert conversion Breath；

(3) for stable state harmonic wave, m-Acetyl chlorophosphonazo signal, directly carry out Hilbert conversion, obtain stable state harmonic wave, The amplitude of m-Acetyl chlorophosphonazo and frequency parameters information；

(4) according to gained amplitude-modulated wave and harmonic wave, m-Acetyl chlorophosphonazo shape information, method of least square is utilized to be fitted, Obtain the digital signal expression formula eventually for electric energy meter test.

EMD decomposition of the present invention i.e. empirical mode decomposition process；To the complexity containing multi-signal composition Waveform, carries out EMD decomposition, obtains the IMF letter of a series of intrinsic mode function, i.e. multi-signal composition Number, meets its maximum and minimizing envelope symmetrical about time shaft, and zero crossing number with greatly, Minimum number at most differs the condition of 1 such that it is able to carries out ensuing Hilbert conversion and calculates its wink Time amplitude and frequency.

Hilbert of the present invention converts, and arbitrary continuous time signal is carried out Hilbert conversion, and signal is each The amplitude of frequency component keeps constant, and 90 ° of displacements occurs in phase place, and after conversion, signal is with original signal composition altogether Yoke plural number is right, it is thus achieved that new analytic signal such that it is able to carry out the extraction of instantaneous amplitude and frequency information.

Method of least square of the present invention, for one group of given data, by retrain all data points with The error sum of squares of match point is minimum, so that it is determined that the parameter value that fitting function model is concrete, obtains matching bent Line.

The invention has the beneficial effects as follows, owing to the most domestic and international test to electric energy meter meter characteristic is all in stable state Carry out under test signal conditioning, test the signal reflection to the stable state included in electrical network Yu Dynamic Signal composition The most comprehensive, therefore, the matching of electric energy meter test signal needs to be more based on electrical network actual operating mode.

And the present invention is by extracting its stable state and Dynamic Signal composition from electrical network measured waveform, including voltage pulsation With harmonic wave, m-Acetyl chlorophosphonazo information, its design parameter of matching obtains digital signal and tests for electric energy meter, it is possible to Make test signal more precisely reflect the practical operation situation of electrical network, electric energy meter dynamic error characteristic is divided Analysis is more accurately rationally.

Accompanying drawing explanation

Fig. 1 is the theory diagram of the inventive method；

Fig. 2 is the EMD decomposition process figure of the present invention；

Fig. 3 is the modulating wave composition that the decomposition of the present invention obtains；

Fig. 4 be the decomposition of the present invention obtain harmonic wave, m-Acetyl chlorophosphonazo composition；

Fig. 5 is multiple amplitude-modulated wave mixed waveform signals of the present invention；

Fig. 6 is the amplitude-modulated wave instantaneous amplitude waveform of the present invention；

Fig. 7 is the instantaneous frequency waveform of the present invention；

Fig. 8 is the harmonic wave of the present invention, m-Acetyl chlorophosphonazo instantaneous amplitude waveform；

Fig. 9 is the harmonic wave of the present invention, m-Acetyl chlorophosphonazo instantaneous frequency waveform.

Detailed description of the invention

The detailed description of the invention of the present invention is as depicted in figs. 1 and 2.Fig. 1 is that the present invention implements schematic diagram, Fig. 2 EMD decomposition process figure for the present invention.

The present embodiment a kind of based on HHT analyze electric energy meter dynamic measuring signal approximating method be embodied as Under:

(1) first electrical network measured waveform signal is carried out EMD decomposition, obtain the table comprised in waveshape signal Levy modulating wave and stable state harmonic wave, the IMF signal of m-Acetyl chlorophosphonazo of dynamic electric voltage fluctuation.

Electrical network actual waveform includes multi-signal composition, including the first-harmonic letter modulated through multiple low frequency amplitude-modulated waves Number and harmonic wave, m-Acetyl chlorophosphonazo signal, waveshape signal form is as follows:

$u\left(t\right)=\[1+\underset{i}{\mathrm{\Σ}}{m}_i\sin \left(2{\mathrm{\πf}}_{i}t\right)\]\sin \left(2{\mathrm{\πf}}_{0}t\right)+\underset{j}{\mathrm{\Σ}}{m}_j\sin \left(2{\mathrm{\πf}}_{j}t\right)$

Wherein, m_iFor the ratio of amplitude-modulated wave amplitude with first-harmonic rated voltage amplitude, f_iFor amplitude-modulated wave frequency, f₀For base Ripple power frequency, m_jFor harmonic wave, the ratio of m-Acetyl chlorophosphonazo amplitude and first-harmonic rated voltage amplitude, f_jFor harmonic wave, m-Acetyl chlorophosphonazo Frequency.

The IMF component processes that EMD decomposes extraction waveshape signal is specific as follows:

First mixed signal u (t) maximum u is asked for₁(t) and minimum u₂The average of (t):

$n=\frac{1}{2}\left(u_1\right(t)+u_2(t\left)\right)$

Taking its difference with u (t) is:

H=u (t)-n

Substituting u (t) with h and repeat said process, until h meets the feature of IMF function, i.e. maximum is with minimum The envelope of value is symmetrical about time shaft, and zero crossing number at most differs 1 with very big, minimum number Condition, now h i.e. can be considered first IMF signal c₁。

Take original signal u (t) and c₁Difference be:

R=u (t)-c₁

Substitute u (t) by difference r and proceed above operation, by that analogy, obtain multiple IMF signal c₂、c₃Deng Deng, iteration to r is that monotone decreasing trend can stop, and now has:

$u\left(t\right)=\underset{i}{\Σ}{c}_i+r$

The final multiple IMF component of signals decomposing acquisition mixed signal u (t), i.e. frequency is the modulation of first-harmonic power frequency Ripple signalWith multiple harmonic waves, m-Acetyl chlorophosphonazo signal

Fig. 3 show and decomposes the modulating wave composition obtained；Fig. 4 show and decomposes harmonic wave, the m-Acetyl chlorophosphonazo one-tenth obtained Point.

(2) for the modulation wave signal characterizing dynamic electric voltage fluctuationTo it Carrying out Hilbert conversion, after conversion, signal and original signal form a conjugate complex number pair, obtain new analytic signal.

Hilbert for continuous time signal X (t) converts specific as follows:

$Y\left(t\right)=\frac{1}{\mathrm{\π}}\underset{-\mathrm{\∞}}{\overset{+\mathrm{\∞}}{\∫}}\frac{X\left(\mathrm{\τ}\right)}{t-\mathrm{\τ}}d\mathrm{\τ}$

Take X (t) and Y (t) to combine, available new conjugated complex signal:

Z (t)=X (t)+jY (t)=a (t) e^iθ(t)

Wherein, a (t), θ (t) respectively its instantaneous amplitude and phase information:

$\left\{\begin{array}{c}a\left(t\right)=\sqrt{X^2\left(t\right)+Y^2\left(t\right)}\\ \mathrm{\θ}\left(t\right)=arctan\left(\frac{Y\left(t\right)}{X\left(t\right)}\right)\end{array}\right.$

Then instantaneous frequency is as follows:

$f\left(t\right)=\frac{1}{2\mathrm{\π}}\frac{d\mathrm{\θ}\left(t\right)}{dt}$

Therefore, for modulation wave signalAfter carrying out Hilbert conversion, newly The instantaneous amplitude of analytic signal be DC component and multiple amplitude-modulated wave sums:

$a\left(t\right)=1+\underset{i}{\Σ}{m}_{i}sin\left(2{\mathrm{\πf}}_{i}t\right)$

Take instantaneous amplitude a (t), after removing DC component, i.e. can get the mixed signal of multiple amplitude-modulated waveIt is carried out EMD decomposition, process and step 1) identical, each amplitude modulation can be extracted Ripple m_i sin(2πf_iT) IMF signal waveform.After it is carried out Hilbert conversion respectively, according to above-mentioned formula, Obtain instantaneous amplitude and the instantaneous frequency waveform of each amplitude-modulated wave.

Fig. 5 is multiple amplitude-modulated wave mixed waveform signals；Fig. 6 is amplitude-modulated wave instantaneous amplitude waveform；Fig. 7 is amplitude modulation Ripple instantaneous frequency waveform.

(3) for stable state harmonic wave, m-Acetyl chlorophosphonazo signalOwing to being decomposed by EMD Arrive the IMF signal m of each harmonic wave, m-Acetyl chlorophosphonazo composition_jsin(2πf_jT), Hilbert therefore can directly be carried out Conversion, then by the instantaneous amplitude of step (2) and instantaneous frequency formula, it is thus achieved that each harmonic wave, m-Acetyl chlorophosphonazo Instantaneous amplitude and instantaneous frequency waveform.

Fig. 8 is the instantaneous amplitude waveform of harmonic wave, m-Acetyl chlorophosphonazo；Fig. 9 is the instantaneous frequency waveform of harmonic wave, m-Acetyl chlorophosphonazo.

(4) according to above step gained amplitude-modulated wave and harmonic wave, the instantaneous amplitude of m-Acetyl chlorophosphonazo and the ripple of instantaneous frequency Shape information, utilizes method of least square to be fitted, and fit procedure is specific as follows:

First determine the function model of matched curve, due to the waveform of required matching be each amplitude-modulated wave with harmonic wave, The instantaneous amplitude of m-Acetyl chlorophosphonazo and instantaneous frequency, therefore function model i.e. linear function m and f.

Then it is determined by normal equation and solves parameter.Assume the instantaneous amplitude m of matching a certain amplitude-modulated wave msin (2 π ft) Curve, has obtained instantaneous amplitude curve data (t_i,m_i) (i=1,2 ..., n), n is the number of curve data point, then matching Target is:

$I=\underset{i}{\overset{n}{\Σ}}{(m-m_i)}^2=min,(i=1,2,...,n)$

The condition i.e. derivative that above formula meets is 0, it may be assumed that

$\frac{dI}{dm}=2\underset{i}{\overset{n}{\Σ}}(m-m_i)=0,(i=1,2,...,n)$

Can obtain:

$m=\frac{\underset{i}{\overset{n}{\Σ}}{m}_i}{n},(i=1,2,...,n)$

Instantaneous frequency f curve fitting process is ibid.

Final obtain each amplitude-modulated wave characterizing dynamic electric voltage fluctuation and stable state harmonic wave, m-Acetyl chlorophosphonazo instantaneous amplitude with Instantaneous frequency m and f parameter value, obtain being analyzed by electrical network measured waveform, can be used for electric energy meter and dynamically survey The digital signal of examination:

$u\left(t\right)=\[1+\underset{i}{\mathrm{\Σ}}{m}_i\sin \left(2{\mathrm{\πf}}_{i}t\right)\]\sin \left(2{\mathrm{\πf}}_{0}t\right)+\underset{j}{\mathrm{\Σ}}{m}_j\sin \left(2{\mathrm{\πf}}_{j}t\right)$

The present invention is by extracting its stable state and Dynamic Signal composition from electrical network measured waveform, including voltage pulsation With harmonic wave, m-Acetyl chlorophosphonazo information, its design parameter of matching obtains digital signal and tests for electric energy meter, it is possible to Make test signal more precisely reflect the practical operation situation of electrical network, electric energy meter dynamic error characteristic is divided Analysis is more accurately rationally.

Claims (4)

1. the electric energy meter dynamic measuring signal approximating method analyzed based on HHT, it is characterised in that said method comprising the steps of:

(1) actual measurement network wave signal is carried out EMD decomposition, respectively obtain modulating wave and stable state harmonic wave, the m-Acetyl chlorophosphonazo signal characterizing dynamic electric voltage fluctuation, modulation wave signal frequency is fundamental frequency, and stable state harmonic frequency signal is first-harmonic integral multiple, and m-Acetyl chlorophosphonazo signal frequency is that first-harmonic is non-integral multiple；

(2) for the modulation wave signal characterizing dynamic electric voltage fluctuation, it is carried out respectively Hilbert conversion and inverse transformation, obtain new analytic signal, solve its instantaneous amplitude, again instantaneous amplitude signal is carried out EMD decomposition, obtain each amplitude-modulated wave signal, solved amplitude and the frequency parameters information of dynamic electric voltage fluctuation by Hilbert conversion；

(3) for stable state harmonic wave, m-Acetyl chlorophosphonazo signal, directly carry out Hilbert conversion, obtain stable state harmonic wave, the amplitude of m-Acetyl chlorophosphonazo and frequency parameters information；

(4) according to gained amplitude-modulated wave and harmonic wave, m-Acetyl chlorophosphonazo shape information, utilize method of least square to be fitted, obtain the digital signal expression formula eventually for electric energy meter test.

A kind of electric energy meter dynamic measuring signal approximating method analyzed based on HHT, it is characterised in that described EMD decomposition i.e. empirical mode decomposition process；To the complicated wave form containing multi-signal composition, carry out EMD decomposition, obtain a series of intrinsic mode function, the i.e. IMF function of multi-signal composition, meet its maximum symmetrical about time shaft with minimizing envelope, and zero crossing number at most differs the condition of 1 with very big, minimum number such that it is able to carry out ensuing Hilbert conversion and calculate its instantaneous amplitude and frequency.

A kind of electric energy meter dynamic measuring signal approximating method analyzed based on HHT, it is characterized in that, described Hilbert converts, arbitrary continuous time signal is carried out Hilbert conversion, the amplitude of each frequency component of signal keeps constant, and 90 ° of displacements occurs in phase place, and after conversion, signal and original signal form a conjugate complex number pair, obtain new analytic signal such that it is able to carry out the extraction of instantaneous amplitude and frequency information.

A kind of electric energy meter dynamic measuring signal approximating method analyzed based on HHT, it is characterized in that, described method of least square, for one group of given data, minimum with the error sum of squares of match point by retraining all data points, so that it is determined that the parameter value that fitting function model is concrete, obtain matched curve.