Abstract:

This dataset contains supporting numerical simulation results used to validate an algorithm for identifying and tracking bubbles under breaking waves. A pseudo-distance error function allows for the identification of continuous, fragmenting, and coalescing bubbles over successive time frames. The parameter space explored includes the weighting of the error function (position, velocity, and volume of bubbles), the number of grid spaces making up a region of interest over which bubbles are identified, and the size of the time frame over which comparisons are made. Three different three-dimensional example fields are included, as are identified bubble, bubble trajectories, and bubble size spectrum and bubble generation size spectrum. A data description file is included with descriptions of the data and format of files in each directory.

Suggested Citation:

Qiang Gao, Grant B. Deane, Han Liu, Lian Shen. 2020. A robust and accurate technique for Lagrangian tracking of bubbles and oil drops and detecting events and supporting data. Distributed by: GRIIDC, Harte Research Institute, Texas A&M University–Corpus Christi. doi:10.7266/6JYHYC9W

Publications:

Gao, Q., Deane, G. B., Liu, H., & Shen, L. (2021). A robust and accurate technique for Lagrangian tracking of bubbles and detecting fragmentation and coalescence. International Journal of Multiphase Flow, 135, 103523. doi:10.1016/j.ijmultiphaseflow.2020.103523

Purpose:

The data was generated for the validation of a novel bubble/oil tracking method.

Data Parameters and Units:

The data description document details the format for files contained in each directory. Briefly, there are files containing the non-dimensional bubble radius and non-dimensional position (x,y,z) for identified bubbles; files Coalescence.DAT, Continuity.DAT, and Fragmentation.DAT which contain the non-dimensional position (x,y,z) at time t1, position (x,y,z) at time t2, the pseudo-error distance, and the ID assigned to each bubble at time t1 and time t2. The ALPHA directory includes data exploring different weights on the error function; the CRIT directory contains data exploring the effect of changing the size of the region of interest over which bubbles are matched (expressed as a number of grid cells); the DT directory contains data supporting exploration of the time frame over which bubbles can be accurately matched; and the T directory contains data from four different time snapshots. The bubble_trajectory directory contains the trajectory of each bubble identified. The format is non-dimensional position (x,y,z) and non-dimensional bubble radius for each time step the bubble was tracked. The bubble_lifetime directory contains two files: Frag_Coal.DAT details the appearance and disappearance mechanisms of each bubble [2=fragmentation, 3=coalescence, 0=other], and Bubble event.DAT details each lifetime: bubble radius, appearance time, disappearance time, and structure [1=cavity, 0 = bubble]. The directory size_spectrum&generation_size_spectrum contains generation_size_spectrum.dat and size_spectrum.dat. Each has the format radius [mm], spectrum [/um radius increment/m^3]. The bubble_id directories contain the three-dimensional field at the defined time step in HDF5 format. The bubbles can be visualized by plotting the contour of phi=0, which corresponds to the air/water interface. GRIDX.IN, GRIDY.IN, and GRIDZ.IN contain the grid cell positions in each direction. Quantities have been non-dimensionalized using L=0.25m and v = 1.565 m/s.

Methods:

The data were generated by numerical simulation. The lateral boundaries are re-entrant, and the air/water interface is treated with the Coupled Level Set Volume of Fluid (CLSVOF) method, which approximates the interface as a mixture of two fluid states. The model domain is λxλxλ/2, where λ is the wavelength; there are (512,384,256) grid points in the span-wise, cross-span, and vertical directions respectively. Grid cells are uniform in size in the span-wise and cross-span direction, but the vertical grid is stretched, with smaller grid cells closer to the surface (approximately λ/512 at the interface). Vectors of the grid box positions are included.

Provenance and Historical References:

[1] Z. Yang, B.-Q. Deng, L. Shen, Direct numerical simulation of wind turbulence over breaking waves, J. Fluid Mech. 850 (2018) 120–155. https://doi.org/10.1017/jfm.2018.466 [2] Y. Hu, X. Guo, X. Lu, Y. Liu, R. A. Dalrymple, L. Shen, Idealized numerical simulation of breaking water wave propagating over a viscous mud layer, Phys. Fluids 24 (2012) 112104. https://doi.org/10.1063/1.4768199 [3] Y. Liu, Numerical study of strong free surface flow and breaking waves, Ph.D. thesis, Johns Hopkins University, 2013. [4] Z. Yang, X.-H. Lu, X. Guo, Y. Liu, L. Shen, Numerical simulation of sediment suspension and transport under plunging breaking waves, Comput. Fluids 158 (2017) 57–71. https://doi.org/10.1016/j.compfluid.2017.03.014 [5] S. Tang, Z. Yang, C. Liu, Y.-H. Dong, L. Shen, Numerical study on the generation and transport of spume droplets in wind over breaking waves, Atmosphere 8 (2017) 248. https://doi.org/10.3390/atmos8120248

Individual Files: