• Proceedings of the Korean Nuclear Society Conference
• /
• 1998.05a
• /
• pp.457-462
• /
• 1998

유동장에서의 기포거동 정보의 중요성 때문에 이를 정확히 측정하기 위한 실험방법이 여러 가지로 발전해 왔지만 아직까지도 기포분포에 대한 정확한 정보 추출에는 도달하지 못하고 있다. 본 연구에서는 원래 의공학분야에서 새로운 tomography 기술로 연구되고 있는 EIT(Electrical Impedance Tomography) 기술을 2상유동에서의 기포분포 측정방법 개발에 적용하기 위한 기초연구와 기포분포 가시화를 위한 전산실험을 수행하였다. 기포분포 가시화를 위해서는 EIT inverse problem solver로 많이 사용되는 iNR(improved Newton-Raphson) 계열의 EIT 염상복원 프로그램을 본 연구진이 유전알고리즘(Genetic Algorithm)과 fuzzy-based mesh grouping 방법을 추가하여 개선한 영상복원프로그램을 사용하였다. 전산실험 결과 본 영상복원프로그램으로는 12$\times$12의 분해능으로 모사되는 기포분포를 저항률 오차한도 $\pm$1%의 신뢰도로 PC상에서 복원이 가능함을 확인하였다.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.3
• /
• pp.151-162
• /
• 2020

This paper presents an acoustic inversion method for estimating the bubble size distribution. The estimation error of the attenuation coefficient represented by a Fredholm integral equation of the first kind is defined as an objective function, and an optimal solution is found by applying the Levenberg-Marquardt (LM) method. In order to validate the effectiveness of the inversion method, numerical simulations using two types of bubble distribution are performed. In addition, a series of experiments are carried out in a water tank (1.0 m × 0.54 m × 0.6 m), using bubbles generated by three different generators. Images of the distributed bubbles are obtained by a high-speed camera, and the insertion losses of the bubble layer are measured using a source and a hydrophone. The image is post-processed to glance a distribution characteristics of each bubble generator. Finally, the size distribution of bubbles is estimated by applying the inversion method to the measured insertion loss. From the inversion results, it was observed that the number of bubbles increases exponentially as the bubble size decreases, and then increases again after the local peak at 70 ㎛ - 120 ㎛.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.3
• /
• pp.198-207
• /
• 2009

Distribution of cavitation bubbles relative to change of the sound speed and attenuation in the water was estimated using acoustic signal from 20 to 300 kHz in two cases that cavitation bubbles exist and do not exist. To study generation and extinction property of cavitation bubble, bubble distribution was estimated in three cases: change of rotation speed (3000-4000 rpm), surface area of blade ($32-98\;mm^2$) and elapsed time (30-120 sec). As a result, the radii of the generated bubbles ranged from 10 to $60{\mu}m$, and bubble radius of $10-20{\mu}m$ and $20-30{\mu}m$ was accounted for 45 and 25% of the total number of cavitation bubbles, respectively. And generation bubble population correlated closely with the rotating speed of the blades but did not correlate with the surface area of blade. It was observed that 80% of total bubble population disappeared within 2 minutes. Finally, acoustic data of bubble distribution was compared with optical data.

• The Journal of the Acoustical Society of Korea
• /
• v.29 no.4
• /
• pp.251-257
• /
• 2010

This paper deals with the attenuation and dispersion of ultrasound in bubbly liquids. Bubble clouds in liquid are formed by a variety of mechanisms, leading to different bubble sizes and spectra. Our aim is to investigate how bubble sizes and spectra affect the attenuation and dispersion characteristics of bubbly liquids. Especially, we highlight the attenuation and dispersion behaviors of nano-bubbles, which have not been reported elsewhere. Computations show that the attenuation and dispersion characteristics of bubbly liquids depend heavily on the quality factors of constituent bubbles. The present study is expected to facilitate in-depth understanding of sound propagation in bubbly liquids.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.4
• /
• pp.305-312
• /
• 2023

Physics-Informed Neural Network (PINN) is used to invert bubble size distributions from attenuation losses. By considering a linear system for the bubble population inversion, Adaptive Learned Iterative Shrinkage Thresholding Algorithm (Ada-LISTA), which has been solved linear systems in image processing, is used as a neural network architecture in PINN. Furthermore, a regularization based on the linear system is added to a loss function of PINN and it makes a PINN have better generalization by a solution satisfying the bubble physics. To evaluate an uncertainty of bubble estimation, deep ensemble is adopted. 20 Ada-LISTAs with different initial values are trained using the same training dataset. During test with attenuation losses different from those in the training dataset, the bubble size distribution and corresponding uncertainty are indicated by average and variance of 20 estimations, respectively. Deep ensemble Ada-LISTA demonstrate superior performance in inverting bubble size distributions than the conventional convex optimization solver of CVX.

• Geophysics and Geophysical Exploration
• /
• v.26 no.1
• /
• pp.8-17
• /
• 2023

Various sources, such as wind, waves, ships, and gas leaks from the seafloor, forms bubbles in the ocean. Underwater bubbles cause signal scattering, considerably affecting acoustic measurements. This characteristic of bubbles is used to block underwater noise by attenuating the intensity of the propagated signal. Recently, researchers have been studying the large-scale release of methane gas as bubble plumes from the seabed. Understanding the physical properties and distribution of bubble plumes is crucial for studying the relation between leaked methane gas and climate change. Therefore, a water tank experiment was conducted to estimate the distribution of bubble plumes using seismic imaging techniques and acoustic signals obtained from artificially generated bubbles using a bubble generator. Reverse time migration was applied to image the bubble plumes while the acquired acoustic envelope signal was used to effectively estimate bubble distribution. Imaging results were compared with optical camera images to verify the estimated bubble distribution. The water tank experiment confirmed that the proposed system could successfully image the distribution of bubble plumes using reverse time migration and the envelope signal. The experiment showed that the scattering signal of artificial bubble plumes can be used for seismic imaging.

• Korean Chemical Engineering Research
• /
• v.49 no.2
• /
• pp.200-205
• /
• 2011

Axial and radial distributions of bubble holdup were investigated in a slurry bubble column with pilot plant scale(D=1.0 m). Effects of gas velocity, surface tension of continuous liquid medium and solid fraction in the slurry phase on the axial and radial distributions of bubble holdup were examined. The bubble holdup decreased with increasing radial dimensionless distance from the center of the column, while it increased with increasing dimensionless distance in the axial direction from the distributor, in all the cases studied. The radial non-uniformity of bubble holdup increased with increasing gas velocity but decreasing surface tension of liquid medium, while it was not dependent upon the solid fraction in the slurry phase. The axial non-uniformity of bubble holdup increased with increasing gas velocity, but it does not change considerably with variations of liquid surface tension or solid fraction in the slurry phase . The axial and radial distributions of bubble holdup were well correlated in terms of operating variables within this experimental conditions.

• Journal of the Korea Concrete Institute
• /
• v.23 no.2
• /
• pp.151-158
• /
• 2011

Bubbles within the foamed concrete manufactured by pre-foaming method is the main factor which affects the physical properties of foamed concrete such as density, strength, and porosity. Although many researches on foamed concrete have been continuously carried out, insufficient number of researches on the properties related to bubbles in the foamed concrete has been performed except for chemical application related researches. In order to make an optimal foamed concrete, study on the bubble properties must be pursued. In order to effectively implement bubbles in the manufacturing of foamed concrete, the bubble properties must be estimated. In this study, in order to determine the bubble properties, examination of the bubble properties according to types and foaming agent concentration was performed. An foaming agent used for this test were anionic surfactant, rosin, and protein system with the foaming agent concentration range of 0.05~13%. Test parameters considered in the study were foaming rate, foam volume, drainage solution volume, and bubble size. The study results showed that, regardless of foaming agent type, higher concentration of foaming agent showed an increase in the foaming rate. Also, the results showed that concentration of foaming agent affected bubble size, drainage solution volume change, and bubble distributions. With respect to the stability of the bubble, protein foaming agent was better than anion surfactant or rosin foaming agent. With respect to the bubble shape, anion surfactant and rosin formed bubbles had polygon shape where as protein formed bubbles had spherical shape.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.14 no.6
• /
• pp.1365-1373
• /
• 1994

A Monte-Carlo simulation model, developed to predict size distribution of air bubbles in turbulent shear flow, is applied to a laboratory-scale problem. Sensitivity to various numerical and physical parameters of the model is analyzed. Practical applicability of the model is explored through comparisons of results with experimental measurements. Bubble size increases with air-water discharge ratio and friction factor. Bubble size decreases with increasing mean flow velocity, but the total bubble surface area in the aeration region remains fairly constant. The effect on bubble size distribution of the longitudinal length increment in the simulation model is negligible. A larger radial length increment yields more small and large bubbles and fewer in between. Bubble size distribution is significantly affected by its initial distribution and the location of air injection. Collision efficiency is introduced to explain the discrepancy between collisions with and without coalescence.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.2
• /
• pp.174-183
• /
• 2022

Acoustic radiation from a submerged elastic shell with an internal fluid surrounded by the bubble layer is studied with the modal theory. An omni-directional point source located on the center of the internal fluid is used as acoustic noise source. The unknown coefficients of modal solutions are solved using the interface conditions between media. To preserve the stability of the modal solution over wide frequency ranges, the scaled technique of modal solution is used. The bubble layer is modeled with four kinds of bubble distribution; uni-modal distribution, uniform distribution, normal distribution, and power-law distribution, based on the effective medium theory of Commander and Prosperetti. For each bubble distribution, the insertion losses are mainly calculated for the frequency. In addition, the numerical simulations are performed depending in the bubble void fraction, the material property of elastic shell, and the gap between the bubble layer and the elastic shell.