What is it about?

As well known, bubble can generate high-speed jet with huge energy when it collapses, which can cause cavitation erosion on the surface of the fluid machinery. But we found that the direction of bubble jet is closely related to the physical characteristics of the boudnaries. For example, high-speed jet directs to the rigid wall, but away from free surface. So we want to learn relationship between the direction of bubble jet and physical characteristics of the boundaries, and then we can control the bubble jet by choosing different boundaries.

Featured Image

Why is it important?

Two reasons to prove this paper is so important. (1) It provides an alternative method to study Fluid-Structure Interaction (FSI) problems. Because traditonal method, such as numerical simulation, is dififcult to treat FSI due to the data transfer on the fluid-structure interface. (2) We found the quantitive relationship between the direction of bubble jet and physical characteristics of the boundaries. So the finding can guide designers to choose appropriate materials to bulid fluid machineries to resist the loading by bubble jet.


I hope this paper can be spread more widely and profoundly. Because it's the first paper to use deep learning machine to study the bubble dynamics. It will inspire the scientists focusing on the bubble dynamics to try a totally new method to solve the problems that cannot be solved at present.

Dr Xiaojian Ma
Beijing Institute of Technology

Read the Original

This page is a summary of: Application of two-branch deep neural network to predict bubble migration near elastic boundaries, Physics of Fluids, October 2019, American Institute of Physics,
DOI: 10.1063/1.5111620.
You can read the full text:




The following have contributed to this page