CN102401858A - Method for detecting fundamental component and harmonic component of voltage of power grid - Google Patents

Abstract

The invention relates to a detection method of a fundamental wave component and a harmonic component of a grid voltage, which relates to the field of signal processing, and solves the problem of low detection accuracy of the existing voltage fundamental wave component and harmonic component detection methods. The method: according to the LMS algorithm, a notch filter is constructed, and the fundamental wave component and each harmonic component are separated in the grid voltage by using the notch filter, so as to obtain the fundamental wave component and each harmonic component of each phase voltage; the obtained The error signal in the fundamental wave component of the A-phase voltage is used as the feedback of the phase-locked loop, and the phase-locked loop is used to obtain the estimated value of the grid voltage frequency and the estimated value of the phase; according to the obtained fundamental wave component of each phase voltage and each Harmonic components, using the symmetrical component method, detect the positive sequence, negative sequence and zero sequence information of the fundamental wave and each harmonic, and use it as the detection result to complete the detection of the fundamental and harmonic components of the grid voltage. The invention is applicable to the detection of fundamental wave and harmonic component of grid voltage.

Description

The detection method of a kind of line voltage fundametal compoment and harmonic component

Technical field

The present invention relates to the signal Processing field, be specifically related to a kind of detection method of voltage fundamental harmonic component.

Background technology

Amplitude, phase angle and the frequency of line voltage first-harmonic and each harmonic positive sequence, negative phase-sequence, zero-sequence component is the generate electricity by way of merging two or more grid systems necessary information of normal operation such as current transformer, uninterrupted power source, active filter, dynamic electric voltage recovery device, electric system FACTS of regenerative resource.Therefore the separation algorithm of studying line voltage first-harmonic and harmonic component has important practical value.

State of Zhao is bright, Liu Baozhi, Xiao Xiangning etc. are in article " a kind of application of improvement d-q conversion in dynamic electric voltage disturbance identification of not having time delay " (electric power network technique; 2004; 28 (7): the improvement dq converter technique that has proposed no time delay 53-57); Zhang Qingchao, Xiao Yulong are at article " a kind of improved electric voltage temporary drop detecting method " (electrotechnics journal; 2006,21 (2): proposed improved α β change detection method etc. 123-126), these class methods can detect the amplitude and the phase angle of electric voltage dropping more fast.These two kinds of methods are to obtain virtual voltage through differentiate by known voltage, have burr yet the A/D sampling can cause the sinusoidal voltage waveform, cause very mistake for the differentiate result.People such as P.Rodriguez and J.Pou is at document " Decoupled double synchronous reference frame PLL for power converters control " (IEEE Transactions on Power Electronics; 2007; 22 (2): proposed improvement phaselocked loop 584-592) based on two synchronous coordinate systems; Applicable to the first-harmonic positive-negative sequence component and the frequency shift (FS) that detect under the unbalance voltage; But when harmonic content was higher, the output of phaselocked loop can produce bigger concussion, and can not detect harmonic component.M.K.Ghartemani and M.Reza Iravani are at document " Amethod for synchronization of power electronic converters in polluted and variable-frequency environments " (IEEE Transactions on Power Systems; 2004; 19 (3): proposed the enhancement mode phaselocked loop 1263-1270); Its main advantage is exactly the mechanism that has changed phase detector; Have more dirigibility, can detect more information, but transient response is slow excessively.People such as C.-H.Huang and C.-H.Lee is at document " Frequcncy estimation ofdistorted power system signals using a robust algorithm " (IEEE Transactions on Power Delivery; 2008; 23 (1): 41-51) proposed a kind of detection method based on Kalman filtering; Can estimate, but just to single-phase situation positive-negative sequence component that can not estimated voltage simultaneously to amplitude, phase angle, the frequency of voltage.

Summary of the invention

The present invention is in order to solve the low problem of accuracy of detection of existing voltage fundamental harmonic component detection method, thereby the detection method of a kind of line voltage fundametal compoment and harmonic component is provided.

The detection method of a kind of line voltage fundametal compoment and harmonic component, it is realized by following steps:

Step 1, according to LMS algorithm construction trapper, gather the voltage signal of each phase of line voltage, adopt said trapper to separate to gathering each phase voltage signal that obtains, obtain the fundametal compoment and the each harmonic component of every phase voltage;

Step 2, with the feedback of the error signal in the fundametal compoment of the A phase voltage that obtains in the step 1 as phaselocked loop; And adopting phaselocked loop to obtain the estimated value of line voltage frequency and the estimated value of phase place, the estimated value of said line voltage frequency and the estimated value of phase place are used to generate the required reference input vector of LMS algorithm of structure trapper;

The fundametal compoment and the each harmonic component of step 3, every phase voltage of obtaining according to step 1; Adopt symmetrical component method; Detect positive sequence, negative phase-sequence and the zero sequence information of first-harmonic and each harmonic, and, accomplish line voltage first-harmonic harmonic component detection as testing result.

Step 1 is described according to LMS algorithm construction trapper; Gather the voltage signal of each phase of line voltage; Each phase voltage signal that adopts said trapper that collection is obtained separates acquisition fundametal compoment and each harmonic component, obtains the fundametal compoment of every phase voltage and the detailed process of each harmonic component and is:

Every phase line voltage is adopted expression formula:

Expression, in the formula, U _nBe the amplitude of nth harmonic, ω is a fundamental frequency, and t is the time,

Be the initial phase angle of nth harmonic, n is a positive integer, and said first-harmonic phase angle is:

In the formula: the first-harmonic phase angle estimated value that

expression is obtained by phaselocked loop; Δ θ representes first-harmonic phase angle evaluated error, so the phase angle of nth harmonic is:

The expression formula of the single-phase line voltage of phase angle substitution of nth harmonic, put in order:

$u\left(t\right)={\mathrm{<figure-callout\; id="1"\; label="U"\; filenames="HDA0000085164420000022.png,HDA0000085164420000023.png"\; state="\{\{state\}\}">U</figure-callout>}}_1\cos \left(\mathrm{\&Delta;\&theta;}\right)\sin \widehat{\mathrm{\&theta;}}+U_1\sin \left(\mathrm{\&Delta;\&theta;}\right)\cos \widehat{\mathrm{\&theta;}}+$

is adjustment factor in the formula, and said adjustment factor is selected the LMS algorithm picks for use;

Line voltage in above-mentioned is expressed as the form of two inner product of vectors, that is:

$\hat{Y}=W^{T}X$

In the formula; The line voltage u (t) that

estimates based on the trapper of LMS for utilization; W is a weight vector, and its expression formula is:

X is with reference to input vector, and expression formula is:

$X={[\sin \left(\widehat{\mathrm{\&theta;}}\right),\cos \left(\widehat{\mathrm{\&theta;}}\right),...,\sin \left(n\widehat{\mathrm{\&theta;}}\right),\cos \left(n\widehat{\mathrm{\&theta;}}\right),...]}^T$

Described in the step 2 with the feedback of the error signal in the fundametal compoment of the A phase voltage that obtains in the step 1 as phaselocked loop, and the detailed process of estimated value that adopts phaselocked loop to obtain estimated value and the phase place of line voltage frequency is:

The expression formula of the weight vector W that step 1 is obtained is revised as:

W＝[W ₁₁，W ₁₂，…，W _n1，W _n2，…] ^T

In the LMS algorithm, utilize error signal e (n) to pass through formula:

W(n+1)＝W(n)+ηe(n)X(n)

Regulate weight vector W, wherein, η is the study step-length, and 0＜η≤1, and X (n) is with reference to the input signal in the input vector;

Then, according to formula:

W ₁₂＝U ₁sin(Δθ)

Obtain the evaluated error of line voltage phase angle;

Adopt classical second-order PLL control system, the evaluated error of line voltage phase angle is carried out filtering and integration, obtain the estimated value of line voltage frequency and the estimated value of phase place.

The fundametal compoment of the described every phase voltage that obtains according to step 1 of step 3 adopts the concrete grammar of positive sequence, negative phase-sequence and the zero sequence information of symmetrical component method detection first-harmonic and each harmonic to be with the each harmonic component:

For the three phase network voltage V (t) of single-frequency, its expression formula is:

Three-phase sinusoidal signal can resolve into positive sequence, negative phase-sequence and zero-sequence component and, that is:

V(t)＝V ⁺(t)+V ^-(t)+V ⁰(t)

V wherein ⁺(t), V ^-(t), V ⁰(t) represent positive sequence, negative phase-sequence and zero-sequence component respectively, that is:

The relation table of the three phase network voltage V (t) of positive sequence, negative phase-sequence and zero-sequence component and single-frequency is shown as following matrixing form:

P wherein _{90 °}Representative is carried out 90 degree phase shifts to V (t), and X, Y are expressed as:

$X=\frac{1}{3}\left[\begin{array}{ccc}1& -0.5& -0.5\\ -0.5& 1& -0.5\\ -0.5& -0.5& 1\end{array}\right]$

$Y=\frac{1}{2\sqrt{3}}\left[\begin{array}{ccc}0& 1& -1\\ -1& 0& 1\\ 1& -1& 0\end{array}\right]$

After linear matrixing, promptly obtain positive sequence, negative phase-sequence and zero-sequence component by the three-phase asymmetrical component.

Beneficial effect: the present invention can exist under the situation such as non-sinusoidal, asymmetric, phase angle sudden change, amplitude sudden change, frequency shift (FS), random noise at line voltage; Accurately detect positive sequence, negative phase-sequence and the zero sequence information of first-harmonic and each harmonic; Has strong robustness, characteristics such as accuracy of detection height.

Description of drawings

Fig. 1 is based on the trapper schematic diagram of LMS; Fig. 2 is improved single-phase phase-locked loop principle schematic; Fig. 3 is a symmetrical components computing unit principle schematic; Figure is the The general frame of 4 line voltage first-harmonics and harmonic component separation algorithm; Fig. 5 is an electrical network three-phase fault voltage oscillogram; Fig. 6 is a line voltage fundamental positive sequence oscillogram; Fig. 7 is a line voltage first-harmonic negative sequence component oscillogram; Fig. 8 is line voltage fundamental wave zero sequence component waveform figure; Fig. 9 is line voltage 5 subharmonic positive-sequence component oscillograms; Figure 10 is line voltage 5 subharmonic negative sequence component oscillograms; Figure 11 is line voltage 5 subharmonic zero-sequence component oscillograms; Figure 12 is a line voltage the seventh harmonic positive-sequence component oscillogram; Figure 13 is a line voltage the seventh harmonic negative sequence component oscillogram; Figure 14 is a line voltage the seventh harmonic zero-sequence component oscillogram; Figure 15 is a line voltage Frequency Estimation oscillogram.

Embodiment

The detection method of embodiment one, a kind of line voltage fundametal compoment and harmonic component, it is realized by following steps:

Step 1, according to LMS algorithm construction trapper, gather the voltage signal of each phase of line voltage, adopt said trapper to separate to gathering each phase voltage signal that obtains, obtain the fundametal compoment and the each harmonic component of every phase voltage;

Step 2, with the feedback of the error signal in the fundametal compoment of the A phase voltage that obtains in the step 1 as phaselocked loop; And adopting phaselocked loop to obtain the estimated value of line voltage frequency and the estimated value of phase place, the estimated value of said line voltage frequency and the estimated value of phase place are used to generate the required reference input vector of LMS algorithm of structure trapper;

The fundametal compoment and the each harmonic component of step 3, every phase voltage of obtaining according to step 1; Adopt symmetrical component method; Detect positive sequence, negative phase-sequence and the zero sequence information of first-harmonic and each harmonic, and, accomplish line voltage first-harmonic harmonic component detection as testing result.

Detailed process is:

For single-phase line voltage, its expression formula is:

In the formula, U _nBe the amplitude of nth harmonic, ω is a fundamental frequency, and t is the time,

Be the initial phase angle of nth harmonic, said first-harmonic phase angle is:

In the formula: the first-harmonic phase angle estimated value that

expression is obtained by phaselocked loop; Δ θ representes first-harmonic phase angle evaluated error, so the phase angle of nth harmonic is:

The expression formula of the single-phase line voltage of phase angle substitution of nth harmonic, put in order:

$u\left(t\right)={\mathrm{<figure-callout\; id="1"\; label="U"\; filenames="HDA0000085164420000022.png,HDA0000085164420000023.png"\; state="\{\{state\}\}">U</figure-callout>}}_1\cos \left(\mathrm{\&Delta;\&theta;}\right)\sin \widehat{\mathrm{\&theta;}}+U_1\sin \left(\mathrm{\&Delta;\&theta;}\right)\cos \widehat{\mathrm{\&theta;}}+$

is adjustment factor in the formula, and said adjustment factor is selected the LMS algorithm picks for use;

Even from following formula, can find out under the initial phase angle condition of unknown of voltage signal, also can reappear original signal through adjustment factor

.The algorithm of adjustment factor is a lot, and this patent adopts the better lowest mean square of practicality (LMS) algorithm.Be illustrated in figure 1 as the trapper schematic diagram based on LMS, the phase information of single-phase line voltage is imported as a reference, single-phase line voltage u (t) is as original input, and the error of system's output is used as the signal of regulating neuron weights W.This detection algorithm belongs to closed loop correction detection algorithm, and accuracy is than higher, and real-time is good.

Line voltage in above-mentioned is expressed as the form of two inner product of vectors, that is:

$\hat{Y}=W^{T}X---\left(5\right)$

In the formula; The line voltage u (t) that estimates based on the trapper of LMS for utilization, W are that weight vector is with reference to input vector with X:

$X={[\sin \left(\widehat{\mathrm{\&theta;}}\right),\cos \left(\widehat{\mathrm{\&theta;}}\right),...,\sin \left(n\widehat{\mathrm{\&theta;}}\right),\cos \left(n\widehat{\mathrm{\&theta;}}\right),...]}^T---\left(7\right)$

Again represent weight vector W, that is:

W＝[W ₁₁，W ₁₂，…，W _n1，W _n2，…] _T (8)

In the LMS algorithm, utilize error signal e (n) to pass through formula:

W(n+1)＝W(n)+ηe(n)X(n) (9)

Regulate weight vector W, wherein, η is the study step-length, and 0＜η≤1, and X (n) is the network reference input signal;

Then, according to formula:

W ₁₂＝U ₁sin(Δθ) (9)

Obtain the evaluated error of line voltage phase angle;

Adopt classical second-order PLL control system, the evaluated error of line voltage phase angle is carried out filtering and integration, obtain the estimated value of line voltage frequency and the estimated value of phase place;

With the error signal of A phase fundametal compoment feedback, estimate line voltage frequency and phase place by phaselocked loop as phaselocked loop.Concrete method is: the evaluated error W that can be obtained the line voltage phase angle by algorithm that step 1 is carried ₁₂, can know the weights W of fundametal compoment by above derivation ₁₂=U ₁Sin (Δ θ).It is thus clear that in weights, comprised the information of line voltage phase angle evaluated error, so can the phase angle evaluated error be adjusted to zero through the suitable closed-loop control system of design.This patent adopts classical second-order PLL control system.Through the filtering of PI loop filter, what obtain is frequency signal, passes through the integral element as voltage controlled oscillator again the phase angle evaluated error, and output is estimated phase angle at last.

Fundametal compoment and the each harmonic component of the every phase voltage that obtains according to step 1 adopt symmetrical component method to detect positive sequence, negative phase-sequence and the zero sequence information of first-harmonic and each harmonic, the detection of realization line voltage first-harmonic harmonic component.

Fundametal compoment and the each harmonic component of the every phase voltage that obtains according to step 1, adopt the concrete grammar of positive sequence, negative phase-sequence and the zero sequence information of symmetrical component method detection first-harmonic and each harmonic to be:

Utilize symmetrical component method, can obtain positive sequence, negative phase-sequence and the zero sequence information of first-harmonic and harmonic wave.Concrete method is: the above detection method of narrating for single-phase signal amplitude, phase place, frequency.If when the generation electric network fault makes three-phase asymmetric, have the appearance of negative phase-sequence, zero-sequence component in the fundametal compoment, be very important so how further separate positive sequence, negative phase-sequence, zero-sequence component this moment exactly.This patent proposed based on the line voltage first-harmonic of LMS and the separation algorithm of harmonic component; Can be applied to simultaneously in the three-phase system; Detect the first-harmonic and the each harmonic component of three-phase voltage respectively, utilize symmetrical component method can further obtain first-harmonic and each harmonic positive sequence, negative phase-sequence, the zero sequence information of each phase then.Be described in detail the ultimate principle of symmetrical component method below.For the three phase network voltage V (t) of single-frequency, its expression formula is:

Three-phase sinusoidal signal can resolve into positive sequence, negative phase-sequence and zero-sequence component and, that is:

V(t)＝V ⁺(t)+V ^-(t)+V ⁰(t) (11)

V wherein ⁺(t), V ^-(t), V ⁰(t) represent positive sequence, negative phase-sequence and zero-sequence component respectively, that is:

The relation table of the three phase network voltage V (t) of positive sequence, negative phase-sequence and zero-sequence component and single-frequency is shown as following matrixing form:

P wherein _{90 °}Representative is carried out 90 degree phase shifts to V (t).And X, Y are expressed as:

$X=\frac{1}{3}\left[\begin{array}{ccc}1& -0.5& -0.5\\ -0.5& 1& -0.5\\ -0.5& -0.5& 1\end{array}\right]---\left(16\right)$

$Y=\frac{1}{2\sqrt{3}}\left[\begin{array}{ccc}0& 1& -1\\ -1& 0& 1\\ 1& -1& 0\end{array}\right]---\left(17\right)$

After linear matrixing, promptly obtain positive sequence, negative phase-sequence and zero-sequence component by the three-phase asymmetrical component.

Shown in Figure 1 is the trapper schematic diagram based on LMS, and the phase information of line voltage is imported as a reference, and single-phase line voltage is as original input.Make inner product with reference to input vector and weight vector, obtain the estimated value of original input, both subtract each other can adjusted weights error signal e.The present invention adopts the LMS algorithm to realize the renewal of weights, and formula is suc as formula shown in (9).Regulate weights according to above iterative formula, then the output of trapper approaches with reference to input with least mean-square error.Thereby realized detection to first-harmonic in the line voltage and each harmonic.

Fig. 2 is improved single-phase phase-locked loop schematic diagram.The basic structure of LMS-a is as shown in Figure 1 among the figure, and its input signal is the output acquisition of the phase information

of A phase line voltage by phaselocked loop.The output signal is the weights W in the LMS algorithm ₁₂, can know W by above derivation ₁₂=V ₁Sin (Δ θ).It is thus clear that in weights, comprised the information of line voltage phase angle evaluated error, so can the phase angle evaluated error be adjusted to zero through the suitable closed-loop control system of design.This patent adopts classical second-order PLL control system.At first, obtain the estimated value of frequency with the filtering of phase angle evaluated error process PI loop filter.Obtain the estimated value of phase angle then through voltage controlled oscillator.

Fig. 3 is a symmetrical components computing unit schematic diagram, utilizes symmetrical component method, can obtain positive sequence, negative phase-sequence and the zero sequence information of first-harmonic and harmonic wave, is described in detail the ultimate principle of symmetrical component method below.If the three phase network voltage V (t) of single-frequency can be expressed as the form of formula (10), three-phase sinusoidal signal can resolve into positive sequence, negative phase-sequence, zero-sequence component and, shown in formula (11), V wherein ⁺(t), V ^-(t), V ⁰(t) represent positive sequence, negative phase-sequence, zero-sequence component respectively.And can be expressed as the matrixing form suc as formula (15), wherein P with the relation of V (t) _{90 °}Representative is carried out 90 degree phase shifts to V (t).Through above simple matrix linear transformation, can derive positive sequence, negative phase-sequence, zero-sequence component by the three-phase asymmetrical component.

Fig. 4 is the The general frame of line voltage first-harmonic and harmonic component separation algorithm.LMS-a wherein, LMS-b, LMS-c structure are as shown in Figure 1, and reference-input signal is obtained by the output of phaselocked loop, and original input is respectively A phase, B phase, C phase voltage.Can know W by above derivation ₁₂=V ₁Sin (Δ θ).It is thus clear that in weights, comprised the information of line voltage phase angle evaluated error, so can the phase angle evaluated error be adjusted to zero through the suitable closed-loop control system of design.This patent adopts classical second-order PLL control system.At first, obtain the estimated value of frequency, obtain the estimated value of phase angle then through voltage controlled oscillator the filtering of phase angle evaluated error process PI loop filter.Trapper by based on LMS can obtain first-harmonic and harmonic component, is entered into then in the symmetrical components computing unit.Can obtain positive sequence, negative phase-sequence, the zero-sequence component of first-harmonic and harmonic wave through the linear matrix conversion.In three-phase phase-locked loop, any phase voltage generation phase angle or amplitude sudden change all can cause very large deviation to Frequency Estimation.Because phaselocked loop has only been used the information of A phase voltage,, strengthened the robustness of algorithm so when other two phases line voltage generation amplitudes, phase angle sudden change, can not exert an influence to Frequency Estimation.

Fig. 5 to Figure 15 occurs under the uneven situation for line voltage, the simulated effect figure of the inventive method.As shown in Figure 5, fault three-phase voltage only contains the unit positive-sequence component, does not contain harmonic components.And when fault took place, uneven situation appearred in line voltage, and contained higher hamonic wave, and variation has also taken place frequency thereupon.Fig. 6 to Fig. 8 is detected line voltage fundamental positive sequence, negative phase-sequence, zero-sequence component, thus it is clear that can detect each component at about two electrical networks about the cycle, and do not have overshoot, excessively relatively steadily.Fig. 9 to 11, Figure 12 to 14 are detected line voltage higher hamonic wave positive sequence, negative phase-sequence, zero-sequence component, though starting stage overshoot is bigger, approximately just tend to be steady through 0.02 second.Figure 15 is the Frequency Estimation waveform of line voltage.

Claims (4)

1. A detection method of grid voltage fundamental component and harmonic component is characterized in that: it is realized by the following steps: Step 1. Construct a notch filter according to the LMS algorithm, collect the voltage signals of each phase of the grid voltage, and use the notch filter to separate the acquired voltage signals of each phase to obtain the fundamental component and harmonics of each phase voltage weight; Step 2, using the error signal in the fundamental wave component of the A-phase voltage obtained in step 1 as the feedback of the phase-locked loop, and using the phase-locked loop to obtain the estimated value of the grid voltage frequency and the estimated value of the phase, the grid voltage frequency The estimated value of and the estimated value of the phase are used to generate the reference input vector required by the LMS algorithm for constructing the notch filter; Step 3. According to the fundamental wave component and each harmonic component of each phase voltage obtained in step 1, use the symmetrical component method to detect the positive sequence, negative sequence and zero sequence information of the fundamental wave and each harmonic, and use it as the detection result , complete the grid voltage fundamental wave and harmonic component detection. 2. the detection method of a kind of grid voltage fundamental wave component and harmonic component according to claim 1, it is characterized in that step one described according to LMS algorithm construction trap wave filter, gathers the voltage signal of each phase of grid voltage, adopts The notch filter separates the collected phase voltage signals to obtain the fundamental component and each harmonic component, and the specific process for obtaining the fundamental component and each harmonic component of each phase voltage is as follows: The grid voltage of each phase is expressed as: In the formula, U _n is the amplitude of the nth harmonic, ω is the fundamental frequency, t is the time,

is the initial phase angle of the nth harmonic, n is a positive integer, and the phase angle of the fundamental wave is: In the formula: Represents the estimated value of the fundamental phase angle obtained by the phase-locked loop, and Δθ represents the estimation error of the fundamental phase angle, so the phase angle of the nth harmonic is:

Substituting the phase angle of the nth harmonic into the expression of the single-phase grid voltage, we can get: $\mathrm{<span\; class="notranslate">\; u</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}\mathrm{<span\; class="notranslate">\; cos</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&Delta;\&theta;</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; sin</span>}\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}\mathrm{<span\; class="notranslate">\; sin</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&Delta;\&theta;</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; cos</span>}\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}<span\; class="notranslate">\; +</span>$

In the formula

For adjustment coefficient, described adjustment coefficient selects LMS algorithm for selection; Express the grid voltage above in the form of the inner product of two vectors, namely: $\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; Y</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; W</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}\mathrm{<span\; class="notranslate">\; x</span>}$ In the formula,

In order to use the grid voltage u(t) estimated by the notch filter based on LMS, W is the weight vector, and its expression is:

X is the reference input vector, the expression is: $\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; sin</span>}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; cos</span>}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; sin</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; cos</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ]</span>}^{\mathrm{<span\; class="notranslate">\; T</span>}}$ 3. The detection method of a kind of grid voltage fundamental wave component and harmonic component according to claim 2, it is characterized in that the error signal in the fundamental wave component of the A-phase voltage obtained in step 1 described in step 2 As the feedback of the phase-locked loop, the specific process of using the phase-locked loop to obtain the estimated value of the grid voltage frequency and the estimated value of the phase is as follows: Modify the expression of the weight vector W obtained in step 1 as: W=[W ₁₁ , W ₁₂ , . . . , W _n1 , W _n2 , . . . ] ^T In the LMS algorithm, the error signal e(n) is used to pass the formula: W(n+1)=W(n)+ηe(n)X(n) Adjust the weight vector W, wherein, η is the learning step size, and 0<η≤1, X(n) is the input signal in the reference input vector; Then, according to the formula: W ₁₂ =U ₁ sin(Δθ) Obtain the estimation error of the grid voltage phase angle; The classic second-order phase-locked loop control system is used to filter and integrate the estimation error of the grid voltage phase angle to obtain the estimated value of the grid voltage frequency and phase. 4. A method for detecting the fundamental wave component and harmonic component of grid voltage according to claim 3, characterized in that the fundamental wave component and each harmonic component of each phase voltage obtained according to step 1 described in step 3 The specific method of using the symmetrical component method to detect the positive sequence, negative sequence and zero sequence information of the fundamental wave and each harmonic is: For the single-frequency three-phase grid voltage V(t), its expression is: The three-phase sinusoidal signal can be decomposed into the sum of positive sequence, negative sequence and zero sequence components, namely: V(t)＝V ⁺ (t)+V ^- (t)+V ⁰ (t) Among them, V ⁺ (t), V ^- (t), and V ⁰ (t) represent positive sequence, negative sequence and zero sequence components respectively, that is:

The relationship between the positive sequence, negative sequence and zero sequence components and the single frequency three-phase grid voltage V(t) is expressed in the following matrix transformation form:

Among them, P _90° represents a 90-degree phase shift for V(t), and X, Y are expressed as: $\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 3</span>}}\left[\begin{array}{ccc}\mathrm{<span\; class="notranslate">\; 1</span>}& <span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.5</span>}& <span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.5</span>}\\ <span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.5</span>}& \mathrm{<span\; class="notranslate">\; 1</span>}& <span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.5</span>}\\ <span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.5</span>}& <span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.5</span>}& \mathrm{<span\; class="notranslate">\; 1</span>}\end{array}\right]$ $\mathrm{<span\; class="notranslate">\; Y</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}\sqrt{\mathrm{<span\; class="notranslate">\; 3</span>}}}\left[\begin{array}{ccc}\mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 1</span>}& <span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}\\ <span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 1</span>}\\ \mathrm{<span\; class="notranslate">\; 1</span>}& <span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\end{array}\right]$ After the linear matrix transformation, the positive sequence, negative sequence and zero sequence components are obtained from the three-phase asymmetric components.

Images