CN110954601A - Water turbine cavitation state online evaluation method based on rapid envelope spectrum kurtosis - Google Patents

Abstract

The invention relates to a water turbine cavitation state online evaluation method based on rapid envelope spectrum kurtosis, which comprises the following steps: step S1: collecting original cavitation signals of a water turbine; step S2, calculating an optimal band-pass filter according to the kurtosis of the fast envelope spectrum, and step S3, performing band-pass filtering by using the optimal band-pass filter to obtain a shock pulse signal waveform; step S4, adopting digital envelope demodulation to solve the impulse envelope waveform; step S5, the number of impact pulses, the time width of each impact pulse and the amplitude of the pulse can be obtained through the identification of the envelope signal; step S6, integrating and calculating the accumulated pulse impact energy in unit time and counting the impact pulse density in unit time; and step S7, obtaining the evaluation of the cavitation state of the water turbine according to the accumulated pulse impact energy in unit time and the counted impact pulse density in unit time. The invention realizes the evaluation of the development degree of the wheel emptying on line by using the density and the impact energy intensity of the impact pulse.

Description

Water turbine cavitation state online evaluation method based on rapid envelope spectrum kurtosis

Technical Field

The invention relates to the field of water turbine cavitation state evaluation, in particular to a water turbine cavitation state online evaluation method based on rapid envelope spectrum kurtosis.

Background

The water turbine is a key energy conversion unit in the hydroelectric generating set, and the working condition of the water turbine can directly influence the running condition of the whole hydroelectric generating set. Cavitation erosion is an inevitable destructive phenomenon for water turbines. Water flow flows through the surface of the flow passage part, and then cavitation can occur locally and slight cavitation erosion is generated, so that honeycomb-shaped cavitation erosion can be formed, even the cavitation erosion develops to blade perforation or edge drop, and the safe operation of the unit can be seriously threatened. Cavitation is one of the main factors that cause the reduction of the operating efficiency of the water turbine, the structural damage and the shortening of the service life of the unit.

Therefore, the reasonable and accurate cavitation state evaluation method designed according to the cavitation generation mechanism and the signal characteristics has important significance for the safe operation and maintenance decision of the water turbine.

Disclosure of Invention

In view of the above, the present invention provides a method for online evaluating a turbine cavitation state based on a fast envelope spectrum kurtosis, which is based on an ultrasonic cavitation signal measured by a wideband ultrasonic sensor mounted on a water turbine, and implements real-time online evaluation of the turbine cavitation state by performing automatic analysis and detection on real-time cavitation data.

In order to achieve the purpose, the invention adopts the following technical scheme:

a water turbine cavitation state online evaluation method based on rapid envelope spectrum kurtosis comprises the following steps:

step S1: collecting original cavitation signals of a water turbine;

step S2, calculating an optimal band-pass filter according to the kurtosis of the fast envelope spectrum;

step S3, performing band-pass filtering by adopting an optimal band-pass filter according to the original cavitation signal to obtain an impact pulse signal waveform;

step S4, according to the obtained impulse signal waveform, adopting digital envelope demodulation to solve the impulse envelope waveform;

step S5, according to the impact pulse envelope waveform, the number of impact pulses, the time width of each impact pulse and the amplitude of the pulse can be obtained by identifying the envelope signal;

step S6, integrating and calculating the accumulated pulse impact energy in unit time and counting the impact pulse density in unit time;

and step S7, obtaining the evaluation of the cavitation state of the water turbine according to the accumulated pulse impact energy in unit time and the counted impact pulse density in unit time.

Further, step S2 is specifically:

step S21: assume the signals are as follows:

in the formula: h (t, omega) is a time-frequency complex envelope of the analyzed signal x (t) and is obtained by adopting fast Fourier transform calculation;

step S22, according to the order moment definition of the spectrum, the spectral kurtosis is expressed as follows:

in the formula: c_4y(ω)Is the fourth order spectral cumulant of signal y (t), and S (ω) is the spectral moment of the instant;

step S23, adopting rapid spectrum kurtosis algorithm to calculate the kurtosis value of the time domain signal under each frequency band, and corresponding the frequency band B (F) with the maximum kurtosis value_c,ΔB_w) As the optimal frequency band, an optimal band pass filter is obtained.

Further, the fast spectral kurtosis algorithm specifically includes:

step S231, setting initial filtering center frequency F_{i_c}And band pass width Δ B_{i_w}；

Step S232, adopting a mode of '1/3-step two' to gradually layer, decompose and adjust the center frequency and the bandwidth to obtain the enveloping spectrum kurtosis under all band-pass filters, and further obtaining the optimal B (F)_c,ΔB_w)。

Further, the step S6 is specifically:

setting the impact envelope waveform function of the cavitation sound signal as x_e(t), assuming that the ith impact pulse is t ═ t₀Start, t ═ t₁End, define E_piEnergy for the impact pulse:

E_picorresponds to the envelopeWave function x_e(t) at [ t₀,t₁]The area between; e_piThe value of (A) is not only related to the amplitude of the pulse but also to the total temporal width of the pulse, the higher the amplitude of the pulse, the higher E_piThe larger the size; the longer the pulse duration, E_piThe larger the size;

is provided with L_{p_T}The total number of pulses detected by the above method for a given time T ═ Δ T; e_piFor the energy of the ith impact pulse in a given time, then

E_pThe accumulated impact energy value of the cavitation impact pulse in unit time is obtained; l is_pCavitation impact pulse density per unit time.

Further, the accumulated impact energy value of the cavitation impact pulse in the unit time comprises the accumulated impact energy E of the cavitation collapse cavitation of the cavitation bubbles in the liquid in the unit time_{p_l}And cumulative impact energy E per unit time of cavitation collapse occurring in the blade/draft tube wall surface_{p_m}(ii) a The impact pulse density per unit time of cavitation includes the impact pulse density per unit time L of cavitation collapse cavitation of cavitation bubbles occurring in the liquid_{p_l}And the impact pulse density per unit time of cavitation collapse cavitation occurring in the blade/draft tube wall surface_{p_m}。

Further, the idle call state evaluation of the water turbine specifically comprises: if E_{p_l}≥E_{p_l_max}And L is_{p_l}≥L_{p_l_max}If so, alarming is carried out due to the cavitation state of the hollow bubbles in the runner liquid; if E_{p_m}≥E_{p_m_max}And L is_{p_m}≥L_{p_m_max}Then, the cavitation state of the surface cavitation of the runner blade/draft tube is alarmed;

wherein E_{p_l_max}The impact energy, L, is accumulated per unit time for cavitation collapse cavitation occurring in the largest tolerable liquid_{p_l_max}The impact pulse density per unit time of cavitation collapse and cavitation of the cavitation bubbles occurring in the largest tolerable liquid; e_{p_m_max}The impact energy L is accumulated per unit time for the cavitation collapse cavitation occurring on the surface of the largest tolerable blade/draft tube wall_{p_m_max}The impact pulse density per unit time of cavitation collapse and cavitation of the cavitation bubbles generated on the surface of the blade/draft tube wall is the largest tolerable.

Compared with the prior art, the invention has the following beneficial effects:

1. the invention realizes real-time online evaluation of the wheel cavitation state by automatic analysis and detection of real-time cavitation data based on ultrasonic cavitation signals measured by a wideband ultrasonic sensor arranged on a water turbine.

2. The invention adopts a fast envelope spectrum kurtosis and digital envelope demodulation mode method to automatically carry out online real-time impulse recognition on ultrasonic section cavitation signals of the water turbine acquired online in real time and calculate accumulated impulse energy and frequency density by integration, and utilizes the density and the intensity of impulse energy of impulse to realize online evaluation on the development degree of the idling of the runner.

Drawings

FIG. 1 is a flow chart illustrating the evaluation of cavitation status of collapsing cavitation in liquid according to an embodiment of the present invention;

FIG. 2 is a flow chart of cavitation state evaluation of collapsed cavitation on the surface of the blade/draft tube according to an embodiment of the present invention;

FIG. 3 is a graph of a raw cavitation sound signal in accordance with an embodiment of the present invention;

FIG. 4 is a diagram illustrating fast envelope spectral kurtosis calculations according to an embodiment of the present invention;

FIG. 5 is a cavitation waveform signal after filtering with an optimal band pass filter in an embodiment of the present invention;

FIG. 6 is a waveform of an envelope of attack of a cavitation sound signal obtained by digital envelope demodulation in an embodiment of the present invention;

FIG. 7 is a schematic representation of energy integration of a single shock pulse by cavitation envelope waveform in accordance with an embodiment of the present invention.

Detailed Description

The invention is further explained below with reference to the drawings and the embodiments.

Referring to fig. 1, the present invention provides a method for online evaluating cavitation status of a water turbine based on kurtosis of a fast envelope spectrum, comprising the following steps:

step S1: ultrasonic cavitation signals measured by a broadband ultrasonic sensor arranged on the water turbine; the original cavitation signal is a complex signal containing multiple frequency components, and the impact pulse signal is almost difficult to identify by directly using the original cavitation signal. Fig. 3 is a diagram of an original cavitation sound signal, and transient impact signals mixed in various frequency components are hardly detected.

Step S2, calculating an optimal band-pass filter according to the kurtosis of the fast envelope spectrum;

step S3, according to the original cavitation signal, performing band-pass filtering by using an optimal band-pass filter to obtain an impact pulse signal waveform, as shown in FIG. 5;

step S4, after obtaining the impulse waveform signal, obtaining the envelope waveform of the impulse signal by using digital envelope demodulation techniques such as Hilbert (H i l bert) transform and the like, as shown in FIG. 6;

step S5, according to the impact pulse envelope waveform, the number of impact pulses, the time width of each impact pulse and the amplitude of the pulse can be obtained by identifying the envelope signal;

step S6, integrating and calculating the accumulated pulse impact energy in unit time and counting the impact pulse density in unit time;

and step S7, obtaining the evaluation of the cavitation state of the water turbine according to the accumulated pulse impact energy in unit time and the counted impact pulse density in unit time.

In this embodiment, the step S2 specifically includes:

step S21: assume the signals are as follows:

in the formula: h (t, omega) is a time-frequency complex envelope of the analyzed signal x (t) and is obtained by adopting fast Fourier transform calculation;

step S22, according to the order moment definition of the spectrum, the spectral kurtosis is expressed as follows:

in the formula: c_4y(ω)Is the fourth order spectral cumulant of signal y (t), and S (ω) is the spectral moment of the instant;

from a signal processing point of view, spectral kurtosis can be understood as the kurtosis value calculated at frequency ω from the output of an ideal filter bank. For each transient signal, an optimal frequency band B (F) is associated_c,ΔB_w) In this frequency band, the kurtosis of this transient signal is greatest.

Therefore, in the actual analysis calculation process, if a frequency band B (F) is found_c,ΔB_w) In this frequency band, the kurtosis value reaches a maximum value, i.e. information about the transient signal can be found, and in this frequency band B (F)_c,ΔB_w) Is the optimal band pass filter that corresponds to its kurtosis at its maximum.

Step S23, adopting rapid spectrum kurtosis algorithm to calculate the kurtosis value of the time domain signal under each frequency band, and corresponding the frequency band B (F) with the maximum kurtosis value_c,ΔB_w) As the optimal frequency band, an optimal band pass filter is obtained.

In this embodiment, the fast spectral kurtosis algorithm specifically includes:

step S231, setting initial filtering center frequency F_{i_c}And band pass width Δ B_{i_w}；

Step S232, adopting a mode of '1/3-step two' to gradually layer, decompose and adjust the center frequency and the bandwidth to obtain the enveloping spectrum kurtosis under all band-pass filters, and further obtaining the optimal B (F)_c,ΔB_w)。

In this embodiment, in order to evaluate cavitation occurring at different positions, it is necessary to calculate an optimal band pass filter corresponding to a cavitation signal of cavitation collapse generated in liquid and an optimal band pass filter corresponding to a cavitation signal of cavitation collapse generated in metal solid (such as a blade and a draft tube wall) respectively, and to initiate the calculationCenter frequency F of filtering_{i_c}And band pass width Δ B_{i_w}Different settings are made. Wherein cavitation caused by cavitation collapse in liquid is evaluated and set

ΔB_{i_w}The cavitation collapse was evaluated at 140 to 30(kHz) on the surface of the blade and the draft tube wall, and the setting was such that

ΔB_{i_w}＝1000-140(kHz)。

FIG. 4 is a diagram illustrating the envelope kurtosis corresponding to the band-pass bands of each hierarchical decomposition computed using the fast envelope spectral kurtosis, where different colors represent different envelope kurtosis values, and the red-colored envelope kurtosis values are larger. It can be intuitively observed from the figure that the envelope kurtosis value reaches a maximum when the band-pass filter is selected at 66666.664Hz,75000.00Hz, and thus 66666.664Hz,75000.00Hz is the optimal band-pass filter for the cavitation signal.

In this embodiment, the step S6 specifically includes:

setting the impact envelope waveform function of the cavitation sound signal as x_e(t), assuming that the ith impact pulse is t ═ t₀Start, t ═ t₁End, define E_piEnergy for the impact pulse:

E_picorresponding to the envelope waveform function x_e(t) at [ t₀,t₁]The area between; e_piThe value of (A) is not only related to the amplitude of the pulse but also to the total temporal width of the pulse, the higher the amplitude of the pulse, the higher E_piThe larger the size; the longer the pulse duration, E_piThe larger the size;

is provided with L_{p_T}The total number of pulses detected by the above method for a given time T ═ Δ T; e_piFor the energy of the ith impact pulse in a given time, then

E_pThe accumulated impact energy value of the cavitation impact pulse in unit time is obtained; l is_pCavitation impact pulse density per unit time.

In the embodiment, according to the mechanism of the cavitation of the rotating wheel, when the cavitation occurs, the cavitation collapse will generate a large amount of random impact pulses; the density L of the impact pulse occurrences can thus be extracted from the measured cavitation sound signal_pAnd the cumulative impact energy E per unit time_pThe evaluation of the cavitation state of the runner can be realized through the two evaluation indexes.

On the one hand, the higher the frequency density L of the impact pulses of cavitation collapse_pIt is explained that the more cavitation bubbles collapse, the more damage to the rotor blade, the draft tube, and the like. On the other hand, the cumulative total impact energy E per unit time_pThe larger the cavitation and the larger the amplitude of the pulse, or the wider the duration of the pulse, the greater the damage to the turbine flow passage components. Thus, by means of the cumulative impact energy E per unit time_pAnd the frequency L of the impact pulses_pThe analysis and evaluation of the cavitation state and development level of the water wheel can be realized.

In this scheme, the following four parameters need to be calculated respectively:

wherein E_{p_l}And L_{p_l}To evaluate the extent of development of cavitation-collapsing cavitation occurring in liquids, E_{p_m}And L_{p_m}To evaluate the extent of development of cavitation collapse occurring at the vane and draft tube wall surfaces.

In the present embodiment, the pair of water turbines has a null-line shapeThe state evaluation specifically comprises: if E_{p_l}≥E_{p_l_max}And L is_{p_l}≥L_{p_l_max}If so, alarming is carried out due to the cavitation state of the hollow bubbles in the runner liquid; if E_{p_m}≥E_{p_m_max}And L is_{p_m}≥L_{p_m_max}Then, the cavitation state of the surface cavitation of the runner blade/draft tube is alarmed;

wherein E_{p_l_max}The impact energy, L, is accumulated per unit time for cavitation collapse cavitation occurring in the largest tolerable liquid_{p_l_max}The impact pulse density per unit time of cavitation collapse and cavitation of the cavitation bubbles occurring in the largest tolerable liquid; e_{p_m_max}The impact energy L is accumulated per unit time for the cavitation collapse cavitation occurring on the surface of the largest tolerable blade/draft tube wall_{p_m_max}The impact pulse density per unit time of cavitation collapse and cavitation of the cavitation bubbles generated on the surface of the blade/draft tube wall is the largest tolerable.

The above description is only a preferred embodiment of the present invention, and all equivalent changes and modifications made in accordance with the claims of the present invention should be covered by the present invention.

Claims (6)

1. A water turbine cavitation state online evaluation method based on rapid envelope spectrum kurtosis is characterized by comprising the following steps:

step S1: collecting original cavitation signals of a water turbine;

step S2, calculating an optimal band-pass filter according to the kurtosis of the fast envelope spectrum;

step S3, performing band-pass filtering by adopting an optimal band-pass filter according to the original cavitation signal to obtain an impact pulse signal waveform;

step S4, according to the obtained impulse signal waveform, adopting digital envelope demodulation to solve the impulse envelope waveform;

step S5, according to the impact pulse envelope waveform, the number of impact pulses, the time width of each impact pulse and the amplitude of the pulse can be obtained by identifying the envelope signal;

step S6, integrating and calculating the accumulated pulse impact energy in unit time and counting the impact pulse density in unit time;

and step S7, obtaining the evaluation of the cavitation state of the water turbine according to the accumulated pulse impact energy in unit time and the counted impact pulse density in unit time.

2. The on-line evaluation method for the cavitation condition of the water turbine based on the kurtosis of the fast envelope spectrum of claim 1, wherein the step S2 is specifically as follows:

step S21: assume the signals are as follows:

in the formula: h (t, omega) is a time-frequency complex envelope of the analyzed signal x (t) and is obtained by adopting fast Fourier transform calculation;

step S22, according to the order moment definition of the spectrum, the spectral kurtosis is expressed as follows:

in the formula: c_4y(ω)Is the fourth order spectral cumulant of signal y (t), and S (ω) is the spectral moment of the instant;

step S23, adopting rapid spectrum kurtosis algorithm to calculate the kurtosis value of the time domain signal under each frequency band, and corresponding the frequency band B (F) with the maximum kurtosis value_c,ΔB_w) As the optimal frequency band, an optimal band pass filter is obtained.

3. The method for on-line evaluation of the cavitation state of a water turbine based on the kurtosis of a fast envelope spectrum as claimed in claim 2, wherein the fast spectral kurtosis algorithm is specifically:

step S231, setting initial filtering center frequency F_{i_c}And band pass width Δ B_{i_w}；

Step S232, adopting a mode of '1/3-step two' to gradually layer, decompose and adjust the center frequency and the bandwidth to obtain the enveloping spectrum kurtosis under all band-pass filters, and further obtaining the optimal B (F)_c,ΔB_w)。

4. The method for on-line evaluation of turbine cavitation based on fast envelope spectral kurtosis as claimed in claim 1, wherein said step S6 specifically comprises:

setting the impact envelope waveform function of the cavitation sound signal as x_e(t), assuming that the ith impact pulse is t ═ t₀Start, t ═ t₁End, define E_piEnergy for the impact pulse:

E_picorresponding to the envelope waveform function x_e(t) at [ t₀,t₁]The area between; e_piThe value of (A) is not only related to the amplitude of the pulse but also to the total temporal width of the pulse, the higher the amplitude of the pulse, the higher E_piThe larger the size; the longer the pulse duration, E_piThe larger the size;

is provided with L_{p_T}The total number of pulses detected by the above method for a given time T ═ Δ T; e_piFor the energy of the ith impact pulse in a given time, then

E_pThe accumulated impact energy value of the cavitation impact pulse in unit time is obtained; l is_pCavitation impact pulse density per unit time.

5. The on-line evaluation method for the cavitation state of the water turbine based on the kurtosis of the fast envelope spectrum as claimed in claim 4, wherein: the accumulated impact energy value of the cavitation impact pulse in the unit time comprises the accumulated impact energy E of cavitation collapse and cavitation in the liquid in the unit time_{p_l}And cumulative impact energy E per unit time of cavitation collapse occurring in the blade/draft tube wall surface_{p_m}(ii) a The density of cavitation impact pulses per unit time includesImpact pulse density per unit time of cavitation collapse cavitation occurring in liquid_{p_l}And the impact pulse density per unit time of cavitation collapse cavitation occurring in the blade/draft tube wall surface_{p_m}。

6. The method for on-line evaluation of the cavitation condition of the water turbine based on the kurtosis of the fast envelope spectrum according to claim 5, wherein the evaluation of the idle-call condition of the water turbine is specifically as follows: if E_{p_l}≥E_{p_l_max}And L is_{p_l}≥L_{p_l_max}If so, alarming is carried out due to the cavitation state of the hollow bubbles in the runner liquid; if E_{p_m}≥E_{p_m_max}And L is_{p_m}≥L_{p_m_max}Then, the cavitation state of the surface cavitation of the runner blade/draft tube is alarmed;

wherein E_{p_l_max}The impact energy, L, is accumulated per unit time for cavitation collapse cavitation occurring in the largest tolerable liquid_{p_l_max}The impact pulse density per unit time of cavitation collapse and cavitation of the cavitation bubbles occurring in the largest tolerable liquid; e_{p_m_max}The impact energy L is accumulated per unit time for the cavitation collapse cavitation occurring on the surface of the largest tolerable blade/draft tube wall_{p_m_max}The impact pulse density per unit time of cavitation collapse and cavitation of the cavitation bubbles generated on the surface of the blade/draft tube wall is the largest tolerable.

Images