A cavitation noise extraction method based on fence effect resampling

Abstract

This invention discloses a cavitation noise extraction method based on picket fence effect resampling. The method includes the following steps: acquiring the acoustic signal segment p of the ultrasonic cavitation sound field. i (t); for the acoustic signal segment p of the ultrasonic cavitation sound field i Perform a Fourier transform on (t) to obtain the signal spectrum p. i (f); Based on the spectral components of the signal spectrum, the sampling rate f of the picket fence effect resampling is obtained through analysis. s‑2 The acoustic signal of the ultrasonic cavitation sound field was resampled using the fence effect to obtain the fence effect resampled signal p. a g ain‑i (t); resampled signal p for the fence effect again‑i (t) After Fourier transform, the 0Hz component is set to zero to obtain the cavitation noise signal spectrum p. cavitation‑i (f); p of the cavitation noise signal spectrum cavitation‑i (f) Perform an inverse Fourier transform to obtain the cavitation noise signal p cavitation‑i (t); Multiple cavitation noise signals are spliced ​​together in the time domain to obtain the total cavitation noise signal. This method is simple to operate and can effectively extract cavitation noise signals from ultrasonic cavitation sound fields.

Description

Technical Field

[0001] This invention belongs to the field of ultrasonic cavitation detection technology, specifically relating to a cavitation noise extraction method based on grating effect resampling, which can be used in the fields of cavitation noise extraction and ultrasonic cavitation research. Background Technology

[0002] The air cavities and bubbles formed under the action of an alternating pressure field (sound field) are called acoustic cavitation, and these air cavities and bubbles are called cavitation bubbles. Acoustic cavitation can include a single cavitation bubble or a group of moving cavitation bubbles excited by the sound field. Based on the lifetime and radius change rate of cavitation bubbles in the liquid medium, acoustic cavitation is mainly of two types: transient cavitation and steady-state cavitation. Steady-state cavitation bubbles exist for a long time, dispersed individually in the water tank. In the acoustic field environment of an ultrasonic cleaning tank, after bubble formation, most oscillate at a fixed position with small amplitude. A small number of bubbles oscillate with larger amplitudes, or collapse after moving to the boundary in the sound field. Transient cavitation is related to the cavitation effect and can produce extreme physical conditions and jet phenomena in the liquid medium. Transient cavitation is included in all common acoustic cavitation phenomena.

[0003] Cavitation noise arises from the linear and nonlinear pulsations of cavitation bubbles, their formation and collapse, and the interactions between them, and is a constant companion to the cavitation process. Under high-power ultrasound, numerous cavitation bubbles are generated in a compact liquid medium, exhibiting large-amplitude pulsations. These large-amplitude pulsations are highly nonlinear, resulting in a highly complex acoustic spectrum of radiated noise, including harmonics, subharmonics, and superharmonics. The collapse of cavitation bubbles is more complex than steady-state pulsations, leading to even more complex spectral components. Steady-state cavitation, due to bubble oscillation and the expansion and contraction of bubble walls, creates microflows around the bubbles. Simultaneously, cavitation noise contains harmonics and subharmonics; with increasing cavitation intensity, superharmonics also appear. Transient cavitation typically occurs in clusters, with bubbles rapidly collapsing after formation, repeating cyclically to form bubble clouds within a certain region. When transient cavitation occurs, broadband noise is generated. Cavitation noise must contain harmonics, subharmonics, superharmonics, and a continuous spectrum that symbolizes the existence of transient cavitation.

[0004] The ultrasonic cavitation sound field contains the fundamental ultrasonic signal, harmonics and subharmonics generated by the nonlinear operation of the transducer, and subharmonics and superharmonics generated by cavitation activity, forming a complex mixed sound field. The fundamental frequency, harmonics, subharmonics, and superharmonics all exist as line spectra in the frequency spectrum. The frequencies of the harmonics, subharmonics, and superharmonics are all based on the fundamental frequency and are multiples of the fundamental frequency (m / n, where m and n are non-zero positive integers), belonging to ordered signals. The disordered noise signal generated by transient cavitation appears as a continuous spectrum.

[0005] The picket fence effect describes the phenomenon where signal sampling only captures information from the sampled points, ignoring data within the sampling interval. It's like viewing a landscape through a fence, only seeing a portion of the spectrum and missing other frequency points, potentially causing some useful frequency components to be missed. Typically, if the "blocked" or lost frequency components are important or characteristic, rendering signal processing meaningless, then if these "blocked" or lost frequency components happen to be the parts that need to be filtered out, they can be utilized.

[0006] There are many types of designs for cavitation noise research and noise extraction, such as the characteristic differentiation between cavitation noise and marine background noise [CN113887482A], cavitation noise separation during ultrasonic solidification [CN114783461A], cavitation intensity measurement of ultrasonic cleaning transducers [CN114623924A], and rotor blade-vortex interference noise extraction [CN115186221A]. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of existing technologies and, based on the characteristics of ultrasonic cavitation and cavitation noise, to provide a method for extracting cavitation noise. This method is simple to operate and can effectively extract cavitation noise signals from ultrasonic cavitation sound fields.

[0008] The technical solution provided by this invention is as follows:

[0009] A cavitation noise extraction method based on picket fence effect resampling, the method comprising the following steps:

[0010] Acquiring the acoustic signal segment p of the ultrasonic cavitation sound field i (t);

[0011] The acoustic signal segment p of the ultrasonic cavitation sound field i Perform a Fourier transform on (t) to obtain the signal spectrum p. i (f);

[0012] Based on the spectral components of the signal spectrum, the sampling rate f of the picket fence effect resampling is obtained through analysis. s-2 The acoustic signal of the ultrasonic cavitation sound field was resampled using the fence effect to obtain the fence effect resampled signal p. again-i (t);

[0013] Resampled signal p due to the picket fence effect again-i (t) After Fourier transform, the 0Hz component is set to zero to obtain the cavitation noise signal spectrum p. cavitation-i (f);

[0014] spectral density of cavitation noise signal p cavitation-i (f) Perform an inverse Fourier transform to obtain the cavitation noise signal p cavitation-i (t);

[0015] Multiple cavitation noise signals are spliced ​​together in the time domain to obtain the total cavitation noise signal.

[0016] Preferably, acquiring the acoustic signal segment of the ultrasonic acoustic field includes the following steps:

[0017] With the ultrasonic transducer working stably, the hydrophone is placed at the measurement position in the sound field, and the received signal measured by the hydrophone is continuously acquired at a preset sampling rate.

[0018] The received signal is segmented according to the preset analysis time to obtain the acoustic signal segments of the ultrasonic cavitation sound field.

[0019] Preferably, the measurement location is a measurement point or a scanning measurement area.

[0020] Preferably, when scanning a measurement area, the scanning should be slow and uniform, and the scanning interval should be less than [a certain value]. λ is the wavelength that drives the ultrasound.

[0021] Preferably, the sampling rate of the fence effect resampling is (1 / n)f0, where f0 is the fundamental frequency of the driving ultrasound and n represents the number of subharmonics to be filtered out.

[0022] Preferably, the expression for the Fourier transform of the resampled signal due to the picket fence effect is:

[0023]

[0024] Where j represents an imaginary number and f represents the signal frequency.

[0025] Preferably, the expression for performing an inverse Fourier transform on the cavitation noise signal spectrum is:

[0026]

[0027] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0028] 1. This method is simple to operate and easy to implement.

[0029] 2. The sampling rate of the fence effect resampling is cleverly set by utilizing the characteristics of the fence effect, ultrasonic cavitation sound field and cavitation noise.

[0030] 3. The cavitation noise signal was effectively extracted through resampling using the picket fence effect.

[0031] 4. The extracted and spliced ​​cavitation noise signal has the same signal structure as the noise signal components in the original signal. Attached Figure Description

[0032] Figure 1Flowchart for resampling cavitation noise extraction for the fence effect.

[0033] Figure 2 The diagram shows the original signal, (a) is a time-domain signal diagram, and (b) is a spectrum diagram.

[0034] Figure 3 The extracted noise signal is shown in (a) as a time-domain signal diagram and (b) as a comparison diagram of its spectrum with the original signal. Detailed Implementation

[0035] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of the present invention.

[0036] A cavitation noise extraction method based on the picket fence effect resampling includes two parts: ultrasonic cavitation noise acquisition and ultrasonic cavitation noise extraction. (Refer to...) Figure 1 As shown, specifically, the method includes the following steps.

[0037] (1) Obtain the acoustic signal segment p of the ultrasonic cavitation sound field i (t), the acoustic signal segment p i The acquisition of (t) includes the following two sub-processes.

[0038] (1-1) With the ultrasonic transducer working stably, place the hydrophone at the measurement position in the sound field for measurement;

[0039] (1-2) The received signal p(t) measured by the hydrophone is continuously acquired at a preset sampling rate, and the sampled signal is segmented by setting an appropriate analysis time to obtain the acoustic signal p of the ultrasonic cavitation sound field. i (t), such as Figure 2 As shown in (a).

[0040] In this invention (1-1), the ultrasonic transducer operates at high power in a liquid medium, exciting cavitation activity to form an ultrasonic cavitation sound field. This field includes ultrasonic transducer radiated sound, noise generated by cavitation activity, etc., and is a multi-component composite sound field. The acoustic signal of the ultrasonic cavitation sound field is a multi-component composite sound signal, containing ordered signal components and disordered noise components. The ordered signal includes the fundamental wave generated by driving the ultrasound, the nonlinear operation of the ultrasonic transducer, and the harmonics generated by cavitation. Its spectrum is in the form of a line spectrum and is affected by the sampling rate during acquisition. The disordered noise signal consists of disordered noise generated by cavitation activity. Its spectrum is a continuous spectrum and its sampling is not affected by the sampling rate or sampling duration.

[0041] In this invention (1-1), the hydrophone can be placed at a suitable point in the sound field for measurement, or a suitable area can be set for scanning measurement. When a suitable area is set for scanning measurement, the scanning should be performed slowly and at a uniform speed, and the scanning interval should be less than [a certain value]. λ is the wavelength at which the ultrasonic transducer drives the ultrasound.

[0042] In this invention (1-2), the hydrophone needs to have sufficient bandwidth to meet the measurement requirements, sufficient sensitivity within the bandwidth, and sufficient robustness to withstand cavitation corrosion and sound pressure impact during high-power operation of the ultrasonic transducer. The acquisition card used to collect the received signal p(t) for hydrophone measurement needs to have a sufficient sampling rate f. s Specifically, it should be no less than 10 times the fundamental frequency of the driving ultrasound.

[0043] (2) The acoustic signal p of the ultrasonic cavitation sound field i Performing a Fourier transform on the (t) segment yields the signal spectrum p. i (f), such as Figure 2 As shown in (b).

[0044] In this invention (2), the spectral components of the acoustic signal of the ultrasonic cavitation sound field are respectively composed of the fundamental frequency generated by driving the ultrasound, the harmonics generated by the nonlinear operation of the ultrasonic transducer and the cavitation activity, and the continuous spectrum generated by transient cavitation.

[0045] (3) Analyze the signal spectrum p i (f) spectral components, and select an appropriate sampling rate f for picket fence resampling. s-2 The acoustic signal of the ultrasonic cavitation sound field was resampled using the fence effect to obtain the fence effect resampled signal p. again-i (t).

[0046] In this invention (3), the sampling rate f of the picket fence effect resampling s-2 = (1 / n)f0, where f0 is the fundamental frequency of the driving ultrasound, and n represents the order of the subharmonics that need to be filtered out. By setting the sampling rate due to the picket fence effect, resampling is performed so that the fundamental component and all harmonic components in the original signal are superimposed on the 0Hz component in the obtained signal.

[0047] (4) Resample the signal p for the fence effect again-i (t) Perform a Fourier transform according to equation (1) to obtain the spectrum p of the resampled signal. again-i (f) At this point, the fundamental, harmonic, and subharmonic components are all added at 0Hz. Therefore, the spectrum of the resampled signal p again-i Setting the 0Hz component in (f) to zero yields the cavitation noise signal spectrum p. cavitation-i (f).

[0048]

[0049] Where j represents an imaginary number and f represents the signal frequency.

[0050] (5) The cavitation noise signal spectrum p cavitation-i (f) Perform an inverse Fourier transform according to equation (2) to obtain the cavitation noise signal p. cavitation-i (t).

[0051]

[0052] The cavitation noise signal is the extracted noise signal, generated by cavitation activity. It has a single component, no line spectrum component in its spectrum, and its probability density distribution is a quasi-normal distribution. Its signal structure is the same as the disordered noise part of the original signal.

[0053] (6) Concatenate multiple cavitation noise signals in the time domain to obtain the total cavitation noise signal p. cavitation (t), whose signal structure is the same as the noise component in the original signal, can be used instead for subsequent cavitation noise analysis.

[0054] like Figure 3 As shown, (a) is a time-domain schematic diagram of the extracted cavitation noise signal, and (b) is a comparison diagram of the spectrum of the extracted cavitation noise signal and the spectrum of the original signal (the spectrum of the ultrasonic cavitation sound field). The fundamental frequency emitted by the ultrasonic transducer is 24kHz. The spectrum of the extracted cavitation noise signal has good agreement with the continuous spectrum of the original signal in the range of 12.5kHz to 130kHz. This frequency range is related to the corresponding sampling rate of the spliced ​​total cavitation noise signal.

[0055] This invention establishes a sampling method based on the fence effect resampling, which filters out interference from fundamental frequency, harmonics, subharmonics, and superharmonics in the ultrasonic cavitation sound field. At the same time, the cavitation noise extraction method based on the fence effect resampling extracts the cavitation noise in the ultrasonic cavitation sound field, laying the foundation for further research on cavitation.

Claims

1. A cavitation noise extraction method based on picket fence effect resampling, characterized in that, The method includes the following steps: Acquiring the acoustic signal segment p of the ultrasonic cavitation sound field i (t); The acoustic signal segment p of the ultrasonic cavitation sound field i Perform a Fourier transform on (t) to obtain the signal spectrum p. i (f); Based on the spectral components of the signal spectrum, the sampling rate f of the picket fence effect resampling is obtained through analysis. s-2 The acoustic signal of the ultrasonic cavitation sound field was resampled using the fence effect to obtain the fence effect resampled signal p. again-i (t); The sampling rate of the fence effect resampling is (1 / n)f0, where f0 is the fundamental frequency of the driving ultrasound and n represents the order of the subharmonics to be filtered out. Resampled signal p due to the picket fence effect again-i (t) After Fourier transform, the 0Hz component is set to zero to obtain the cavitation noise signal spectrum p. cavitation-i (f); spectral density of cavitation noise signal p cavitation-i (f) Perform an inverse Fourier transform to obtain the cavitation noise signal p cavitation-i (t); Multiple cavitation noise signals are spliced ​​together in the time domain to obtain the total cavitation noise signal.

2. The cavitation noise extraction method based on picket fence effect resampling as described in claim 1, characterized in that, The acquisition of the acoustic signal segment of the ultrasonic cavitation sound field includes the following steps: With the ultrasonic transducer working stably, the hydrophone is placed at the measurement position in the sound field, and the received signal measured by the hydrophone is continuously acquired at a preset sampling rate. The received signal is segmented according to the preset analysis time to obtain the acoustic signal segment of the ultrasonic cavitation sound field.

3. The cavitation noise extraction method based on picket fence effect resampling as described in claim 2, characterized in that, The measurement location is either a measurement point or a scanning measurement area.

4. The cavitation noise extraction method based on picket fence effect resampling as described in claim 3, characterized in that, When scanning a measurement area, the scanning should be slow and uniform, and the scanning interval should be less than [a certain value]. λ is the wavelength of the ultrasonic transducer driving the ultrasound.

5. The cavitation noise extraction method based on picket fence effect resampling as described in claim 1, characterized in that, Resampled signal p due to the picket fence effect again-i The expression for (t) through Fourier transform is: ， Where j represents an imaginary number and f represents the signal frequency.

6. The cavitation noise extraction method based on picket fence effect resampling as described in claim 1, characterized in that, spectral density of cavitation noise signal p cavitation-i (f) The expression for the inverse Fourier transform is: ， Where j represents an imaginary number and f represents the signal frequency.

Images