CN104089778A - Water turbine vibration fault diagnosis method - Google Patents

Abstract

The invention provides a water turbine vibration fault diagnosis method. The water turbine vibration fault diagnosis method is characterized in that the water turbine vibration fault diagnosis method includes the steps that wavelet transformation is used for conducting self-adaptive denoising on collected water turbine vibration signals, and the complexity of data is reduced while noise influence is reduced; a local mean decomposition method is used for conducting self-adaptive decomposition on signals pre-processed through wavelets, and fault feature information in the water turbine vibration signals is extracted; according to the vibration principle of a water turbine, the extracted feature information such as frequency and energy is analyzed, and the operating state of the water turbine is diagnosed. Firstly, wavelet transformation is used for pre-processing the signals, the noise influence on the processed signals is reduced, the complexity of the signals is reduced, the entropy of the signals is reduced, and thus the fault feather information can be extracted more accurately and more effectively through local mean decomposition.

Description

A kind of hydraulic turbine vibrating failure diagnosis method

Technical field

The invention belongs to vibration equipment fault diagnosis technology field, is a kind of hydraulic turbine vibrating failure diagnosis method, more particularly, is a kind of hydraulic turbine vibrating failure diagnosis method based on wavelet pretreatment and the decomposition of local average.

Background technology

Hydraulic turbine operation is in the environment such as very noisy, electromagnetic interference (EMI), while breaking down, generally all can in vibration, embody to some extent, its vibration signal has non-stationary, many modulation, multicomponent feature, adopt classic method, often be difficult to extract accurately the fault characteristic information of the hydraulic turbine, at present, the analyzing and processing means of signal are that one of restriction fault diagnosis technology development hinders greatly.

Fourier transform is a kind of integral transformation, and whole time-domain signal is moved on frequency domain and analyzed, and it lacks the description to signal detail, and failure message often embodies in detail, and Fourier transform lacks the amplification effect to detailed information.

Wavelet transformation, although its time frequency window is variable, but can not obtain higher time domain and frequency domain resolution simultaneously, the essence of wavelet transformation is to remove approximate signal to be analyzed with the linear combination of one group of wavelet basis function, this essence is a kind of linear transformation, when processing nonlinear properties, there is time error very large.

Hilbert-Huang transform is a kind of adaptive Time-Frequency Analysis Method, however its cross envelope, owe envelope, the problem such as end effect, and the edge effect when calculating instantaneous amplitude and instantaneous frequency and the negative frequency problem that produces without physical significance remain unsolved.

Summary of the invention

The object of the invention is to overcome the shortcoming of prior art, a kind of hydraulic turbine vibrating failure diagnosis method is provided, should can effectively extract the fault characteristic information being included in hydraulic turbine vibration signal in this way, and on this basis, the running status of the diagnosis hydraulic turbine.

The object of the invention is to be realized by following technical scheme:

A hydraulic turbine vibrating failure diagnosis method, is characterized in that, it comprises the following steps:

1) adopt wavelet transformation, to the hydraulic turbine vibration signal self-adaptation denoising collecting, when reducing noise effect, reduce the complicacy of data;

2) signal after utilizing local average decomposition method to wavelet pretreatment carries out adaptive decomposition, extracts the fault characteristic information in hydraulic turbine vibration signal;

3), in conjunction with the vibration mechanism of the hydraulic turbine, analyze the characteristic informations such as the frequency extracted, energy, the running status of the diagnosis hydraulic turbine.

It is the hydraulic turbine vibration signal self-adaptation denoising to collecting that described step (1) adopts wavelet transformation, and wherein, the selection course of adaptive threshold is as follows:

(1) n of this layer of decomposition high frequency coefficient D pressed to the sequence of mould value size, set initial threshold, by the following coefficient zero setting of threshold value, D=(D ₁, D ₂... D _k, 0...0), D wherein ₁to D _kfor the wavelet coefficient retaining;

(2) energy of the wavelet coefficient that calculating retains accounts for signal energy percentage η _i=∑ (d _j) ²/ ∑ (d _j) ²+ ∑ (d _i) ²and wavelet coefficient is zero the shared percentage ξ of number _i=(n-k)/n, wherein, d _jfor the wavelet coefficient retaining, d _iwavelet coefficient for zero setting;

(3) select abs (η _i-ξ _i) corresponding minimum coefficient is as threshold value T _i;

(4) do soft-threshold quantification treatment:

$d_i=\left\{\begin{array}{cc}|d_i-T_i|\&CenterDot;\mathrm{sign}\left(d_i\right)& \left|d_i\right|>T_i\\ 0& \left|d_i\right|<T_i\end{array}\right..$

Technique effect of the present invention is: first by wavelet transformation, signal is carried out to pre-service, signal reduction affected by noise after processing, and the reduced complexity of signal, the entropy of signal reduces, through local average, decompose so again extraction fault characteristic information that can be more accurate and effective.

Accompanying drawing explanation

Fig. 1 is a kind of hydraulic turbine vibrating failure diagnosis method of the present invention process flow diagram;

Fig. 2 is hydraulic turbine machine cap horizontal vibration time domain waveform figure;

Fig. 3 is the hydraulic turbine machine cap horizontal vibration oscillogram after wavelet pretreatment;

Fig. 4 is local average decomposition result figure;

Fig. 5 is the time-frequency energy profile being built by each PF component.

Embodiment

With reference to Fig. 1, a kind of hydraulic turbine vibrating failure diagnosis method, comprises the following steps:

1) adopt wavelet transformation, to the hydraulic turbine vibration signal self-adaptation denoising collecting, when reducing noise effect, reduce the complicacy of data;

2) signal after utilizing local average decomposition method to wavelet pretreatment carries out adaptive decomposition, extracts the fault characteristic information in hydraulic turbine vibration signal;

3), in conjunction with the vibration mechanism of the hydraulic turbine, analyze the characteristic informations such as the frequency extracted, energy, the running status of the diagnosis hydraulic turbine.

The hydraulic turbine machine cap of take below vibration is example, and at machine cap, place arranges sensor, and the signal collecting carries out adaptive threshold denoising with db4 small echo, and the selection course of adaptive threshold is as follows:

(1) the high frequency coefficient D of this layer of decomposition is pressed to the sequence of mould value size, set initial threshold, by the following coefficient zero setting of threshold value, k coefficient before retaining, D=(D ₁, D ₂... D _k, 0...0).

(2) energy of calculating nonzero coefficient accounts for signal energy percentage η _i=∑ (d _j) ²/ ∑ (d _j) ²+ ∑ (d _i) ²and wavelet coefficient is zero the shared percentage ξ of number _i, wherein, d _jfor the coefficient retaining, d _icoefficient for zero setting.

(3) select abs (η _i-ξ _i) corresponding minimum coefficient is as threshold value T _i.

(4) soft-threshold quantification treatment:

$d_i=\left\{\begin{array}{cc}|d_i-T_i|\&CenterDot;\mathrm{sign}\left(d_i\right)& \left|d_i\right|>T_i\\ 0& \left|d_i\right|<T_i\end{array}\right.$

Fig. 2 is for collecting hydraulic turbine machine cap horizontal vibration time domain waveform figure, as can be seen from the figure, vibration pack has contained a large amount of noise effects, after the denoising of small echo self-adaptation, as shown in Figure 3, noise effect obviously reduces its vibrational waveform, and the waveform before and after contrast denoising, in trend, have similarity, the feature of known signal is retained.

Signal after wavelet pretreatment is carried out to adaptive decomposition by local average decomposition method, and each component obtaining after decomposition has actual physical meaning, and the characteristic information that has comprised original signal, and decomposable process is as follows:

(1) determine all Local Extremum n of original signal x (t) _i, obtain the mean value m of all adjacent Local Extremum _i:

$m_i=\frac{n_i+n_{i+1}}{2}---\left(1\right)$

By all adjacent mean point m _iwith straight line, couple together, then with moving average method, carry out smoothing processing, obtain local mean value function m ₁₁(t);

(2) obtain envelope estimated value

$a_i=\frac{|n_i-n_{i+1}|}{2}---\left(2\right)$

By all adjacent two envelope estimated value a _iwith straight line, connect, then adopt moving average method to carry out smoothing processing, obtain envelope estimation function a ₁₁(t);

(3) by local mean value function m ₁₁(t) from original signal x (t), separate, obtain removing the h of low frequency signal part ₁₁(t),

h ₁₁(t)＝x(t)-m ₁₁(t) (3)

(4) utilize h ₁₁(t) divided by envelope estimation function a ₁₁(t) with to a ₁₁(t) carry out demodulation, obtain FM Function s ₁₁(t),

s ₁₁(t)＝h ₁₁(t)/a ₁₁(t) (4)

(5) with s ₁₁(t) replace original signal x (t) to repeat above-mentioned steps (1)-(4) and just can obtain s ₁₁(t) envelope estimation function a ₁₂(t), judgement a ₁₂(t) whether equal 1, if equal 1, forward step (6) to, otherwise, with s _1n(t) replace x (t), repeating step (1) is to (4), until s _1n(t) be pure FM signal, i.e. a s _1n(t) envelope estimation function a _{1 (n+1)}(t)=1, in practical engineering application, stopping criterion for iteration can be a _1n(t) ≈ 1, and each component producing in iterative process is as follows,

$\left\{\begin{array}{c}{h}_{11}\left(t\right)=x\left(t\right)-m_{11}\left(t\right)\\ h_{12}\left(t\right)=s_{11}\left(t\right)-m_{12}\left(t\right)\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ h_{1n}\left(t\right)=s_{1(n-1)}\left(t\right)-m_{1n}\left(t\right)\end{array}\right.---\left(5\right)$

$\left\{\begin{array}{c}{s}_{11}\left(t\right)=h_{11}\left(t\right)/a_{11}\left(t\right)\\ s_{12}\left(t\right)=h_{12}\left(t\right)/a_{12}\left(t\right)\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ s_{1n}\left(t\right)=h_{1n}\left(t\right)/a_{1n}\left(t\right)\end{array}\right.---\left(6\right)$

In formula, m _1n(t) be FM Function s _1n(t) local mean value function, a _1n(t) be s _{1 (n-1)}(t) envelope estimation function;

(6) all envelope estimation functions that produce in iterative process can be obtained to envelope signal (instantaneous amplitude function) mutually at convenience

$a_1\left(t\right)=a_{11}\left(t\right)a_{12}\left(t\right)\&CenterDot;\&CenterDot;\&CenterDot;a_{1n}\left(t\right)=\underset{q=1}{\overset{n}{\mathrm{\&Pi;}}}{a}_{1q}\left(t\right)---\left(7\right)$

(7) by envelope signal a ₁and pure FM signal s (t) _1n(t) can obtain at convenience mutually first PF component of original signal:

PF ₁(t)＝a ₁(t)s _1n(t) (8)

It has comprised frequency content the highest in original signal, is the frequency modulation of a simple component---amplitude-modulated signal, and its instantaneous amplitude is exactly envelope signal a ₁(t), its instantaneous frequency f ₁(t) can be by pure FM signal s _1n(t) obtain, that is:

$f_1\left(t\right)=\frac{1}{2\mathrm{\&pi;}}\frac{d\left[\arccos \left(s_{1n}\left(t\right)\right)\right]}{\mathrm{dt}}---\left(9\right)$

(8) by first PF component PF ₁(t) separated from original signal x (t), obtain a new signal u ₁(t);

(9) by u ₁(t) replace x (t) repeating step (1)-(8), circulation k time, until u _k(t) be till a monotonic quantity,

$\left\{\begin{array}{c}{u}_1\left(t\right)=x\left(t\right)-{\mathrm{PF}}_1\left(t\right)\\ u_2\left(t\right)=u_1\left(t\right)-{\mathrm{PF}}_2\left(t\right)\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ u_k\left(t\right)=u_{k-1}\left(t\right)-{\mathrm{PF}}_k\left(t\right)\end{array}\right.---\left(10\right)$

In formula, PF _k(t) for decomposed each PF component obtaining, u by original signal _k(t) be remaining component;

Original signal x (t) can be by all PF component and u _k(t) reconstruct, that is:

$x\left(t\right)=\underset{p=1}{\overset{k}{\mathrm{\&Sigma;}}}{\mathrm{PF}}_p\left(t\right)+u_k\left(t\right)---\left(11\right).$

Referring to Fig. 4, local average decomposition method, that is the decomposition result of LMD, each component of LMD has actual physical meaning, each PF component being obtained by decomposition is asked for instantaneous frequency and the amplitude of signal, build three-dimensional time-frequency energy profile, see Fig. 5, as can be seen from the figure, the unit rotational frequency that has comprised 5hz in the vibration information of the hydraulic turbine, and 1/4 faint frequency multiplication, 2-4 frequency multiplication, and stronger 50hz frequency content.From hydraulic turbine vibration mechanism, 50hz composition is wherein the noise effect that system acquisition data Shi Yin power plant imperfect earth produces, and 1/4 frequency multiplication composition is that the low frequency whirlpool band under this load causes.Through repeatedly analytical test, when guide vane opening is adjusted to after specified aperture, 2-4 frequency multiplication composition disappears, can analyze thus, this 2-4 frequency multiplication composition is because waterpower inequality causes, blade openings that this kind of situation may be stator is unbalanced, linear discontinuities, or because seal ring circularity is inadequate, causes runner to go out stream circumferentially inhomogeneous.This phenomenon can not cause too much influence to the normal operation of the hydraulic turbine, and this load area operation is avoided in suggestion.As can be seen here, the local average decomposition method based on wavelet pretreatment, can extract the characteristic information in hydraulic turbine vibration signal under strong noise background, and can effectively diagnose the running status of unit.

Claims (2)

1. a hydraulic turbine vibrating failure diagnosis method, is characterized in that, it comprises the following steps:

1) adopt wavelet transformation, to the hydraulic turbine vibration signal self-adaptation denoising collecting, when reducing noise effect, reduce the complicacy of data;

2) signal after utilizing local average decomposition method to wavelet pretreatment carries out adaptive decomposition, extracts the fault characteristic information in hydraulic turbine vibration signal;

3), in conjunction with the vibration mechanism of the hydraulic turbine, analyze the characteristic informations such as the frequency extracted, energy, the running status of the diagnosis hydraulic turbine.

2. a kind of hydraulic turbine vibrating failure diagnosis method according to claim 1, is characterized in that, it is the hydraulic turbine vibration signal self-adaptation denoising to collecting that described step (1) adopts wavelet transformation, and wherein, the selection course of adaptive threshold is as follows:

(1) n of this layer of decomposition high frequency coefficient D pressed to the sequence of mould value size, set initial threshold, by the following coefficient zero setting of threshold value, D=(D ₁, D ₂... D _k, 0...0), D wherein ₁to D _kfor the wavelet coefficient retaining;

(2) energy of the wavelet coefficient that calculating retains accounts for signal energy percentage η _i=∑ (d _j) ²/ ∑ (d _j) ²+ ∑ (d _i) ²and wavelet coefficient is zero the shared percentage ξ of number _i=(n-k)/n, wherein, d _jfor the wavelet coefficient retaining, d _iwavelet coefficient for zero setting;

(3) select abs (η _i-ξ _i) corresponding minimum coefficient is as threshold value T _i;

(4) do soft-threshold quantification treatment:

$d_i=\left\{\begin{array}{cc}|d_i-T_i|\&CenterDot;\mathrm{sign}\left(d_i\right)& \left|d_i\right|>T_i\\ 0& \left|d_i\right|<T_i\end{array}\right..$