Bubble patterns recognition using neural networks: Application to the analysis of a two-phase bubbly jet

Poletaev, Igor; Tokarev, M. P.; Pervunin, Konstantin S.

doi:10.1016/j.ijmultiphaseflow.2019.103194

Cited by 73 publications

(25 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, there have been publications devoted to the use of deep learning for automatic object recognition in materials science and related fields. For example, a number of studies were aimed at searching for defects in metals [ 12 , 13 , 14 , 15 , 16 ] including images of atomically resolved scanning transmission electron microscopy [ 17 ], classification of objects in scanning electron microscope images [ 18 ], and determining bubbles sizes in thermophysical processes [ 19 ].…”

Section: Introductionmentioning

confidence: 99%

Nanoparticle Recognition on Scanning Probe Microscopy Images Using Computer Vision and Deep Learning

et al. 2020

View full text Add to dashboard Cite

Identifying, counting and measuring particles is an important component of many research studies. Images with particles are usually processed by hand using a software ruler. Automated processing, based on conventional image processing methods (edge detection, segmentation, etc.) are not universal, can only be used on good-quality images and need to set a number of parameters empirically. In this paper, we present results from the application of deep learning to automated recognition of metal nanoparticles deposited on highly oriented pyrolytic graphite on images obtained by scanning tunneling microscopy (STM). We used the Cascade Mask-RCNN neural network. Training was performed on a dataset containing 23 STM images with 5157 nanoparticles. Three images containing 695 nanoparticles were used for verification. As a result, the trained neural network recognized nanoparticles in the verification set with 0.93 precision and 0.78 recall. Predicted contour refining with 2D Gaussian function was a proposed option. The accuracies for mean particle size calculated from predicted contours compared with ground truth were in the range of 0.87–0.99. The results were compared with outcomes from other generally available software, based on conventional image processing methods. The advantages of deep learning methods for automatic particle recognition were clearly demonstrated. We developed a free open-access web service “ParticlesNN” based on the trained neural network, which can be used by any researcher in the world.

show abstract

Section: Introductionmentioning

confidence: 99%

Nanoparticle Recognition on Scanning Probe Microscopy Images Using Computer Vision and Deep Learning

et al. 2020

View full text Add to dashboard Cite

show abstract

“…The figures show that the bubbles in the separation region of the flow can have a substantially non-spherical shape. The currently developed methods, as a rule, deal with spherical or elliptical bubbles [24]. Development of a deep learning-based image processing technique for bubble pattern recognition and shape reconstruction in dense bubbly flows were performed in [25,26].…”

Section: Measurement Setupmentioning

confidence: 99%

The Effect of a Backward-Facing Step on Flow and Heat Transfer in a Polydispersed Upward Bubbly Duct Flow

Bogatko

Chinak

Evdokimenko

et al. 2021

Water

View full text Add to dashboard Cite

The experimental and numerical results on the flow structure and heat transfer in a bubbly polydispersed upward duct flow in a backward-facing step are presented. Measurements of the carrier fluid phase velocity and gas bubbles motion are carried out using the PIV/PLIF system. The set of RANS equations is used for modeling the two-phase bubbly flow. Turbulence of the carrier fluid phase is predicted using the Reynolds stress model. The effect of bubble addition on the mean and turbulent flow structure is taken into account. The motion and heat transfer in a dispersed phase is modeled using the Eulerian approach taking into account bubble break-up and coalescence. The method of delta-functions is employed for simulation of distributions of polydispersed gas bubbles. Small bubbles are presented over the entire duct cross-section and the larger bubbles mainly observed in the shear mixing layer and flow core. The recirculation length in the two-phase bubbly flow is up to two times shorter than in the single-phase flow. The position of the heat transfer maximum is located after the reattachment point. The effect of the gas volumetric flow rate ratios on the flow patterns and maximal value of heat transfer in the two-phase flow is studied numerically. The addition of air bubbles results in a significant increase in heat transfer (up to 75%).

show abstract

“…Here, the auxiliary variable is the interface position, i.e. the volume fraction field, which in an experimental setup is obtained using a combination of optical techniques and image processing (Murai et al, 2001(Murai et al, , 2006Takamasa et al, 2003;Poletaev et al, 2020). Three 2D test cases are investigated, namely a drop in a shear flow, an oscillating drop and a rising bubble.…”

Section: Inverse Problemsmentioning

confidence: 99%

“…Here, the neural network receives data on the interface position scattered over time, motivated by experimental setups where the interface position is obtained using non-invasive methods, namely optical techniques combined with image processing (Murai et al, 2001;Takamasa et al, 2003;Murai et al, 2006;Poletaev et al, 2020), provided a direct observation of the interface is possible. In experiments that do not allow for direct observation, methods such as ultrasonic detection (Murai et al, 2010) may be used.…”

Section: Introductionmentioning

confidence: 99%

Inferring incompressible two-phase flow fields from the interface motion using physics-informed neural networks

Buhendwa¹,

Adami²,

Adams³

2021

Preprint

View full text Add to dashboard Cite

In this work, physics-informed neural networks are applied to incompressible two-phase flow problems. We investigate the forward problem, where the governing equations are solved from initial and boundary conditions, as well as the inverse problem, where continuous velocity and pressure fields are inferred from scattered-time data on the interface position. We employ a volume of fluid approach, i.e. the auxiliary variable here is the volume fraction of the fluids within each phase.For the forward problem, we solve the two-phase Couette and Poiseuille flow. For the inverse problem, three classical test cases for two-phase modeling are investigated: (i) drop in a shear flow, (ii) oscillating drop and (iii) rising bubble. Data of the interface position over time is generated by numerical simulation. An effective way to distribute spatial training points to fit the interface, i.e. the volume fraction field, and the residual points is proposed. Furthermore, we show that appropriate weighting of losses associated with the residual of the partial differential equations is crucial for successful training. The benefit of using adaptive activation functions is evaluated for both the forward and inverse problem.

show abstract

Bubble patterns recognition using neural networks: Application to the analysis of a two-phase bubbly jet

Cited by 73 publications

References 22 publications

Nanoparticle Recognition on Scanning Probe Microscopy Images Using Computer Vision and Deep Learning

Nanoparticle Recognition on Scanning Probe Microscopy Images Using Computer Vision and Deep Learning

The Effect of a Backward-Facing Step on Flow and Heat Transfer in a Polydispersed Upward Bubbly Duct Flow

Inferring incompressible two-phase flow fields from the interface motion using physics-informed neural networks

Contact Info

Product

Resources

About