Abstract:

Hydro3D includes an Eulerian-Lagrangian algorithm for large-eddy simulation of bubble plume dynamics. The main numerical novelties consist on defining the gas-liquid coupling based on the bubble size to mesh resolution ratio where the interpolation between Eulerian and Lagrangian frameworks is achieved through the use of delta functions. The model's performance has been thoroughly validated for bubble plumes in tanks, cross flows and subjected to many different conditions.

Suggested Citation:

Thorsten Stoesser. 2016. HYDRO3D: Large Eddy Simulation (LES) of the near-field evolution of a subsea oil and gas plume. Distributed by: GRIIDC, Harte Research Institute, Texas A&M University–Corpus Christi. doi:10.7266/N76W980Z

Publications:

Fraga, B., Stoesser, T., Lai, C. C. K., & Socolofsky, S. A. (2016). A LES-based Eulerian–Lagrangian approach to predict the dynamics of bubble plumes. Ocean Modelling, 97, 27–36. doi:10.1016/j.ocemod.2015.11.005

Fraga, B., & Stoesser, T. (2016). Influence of bubble size, diffuser width, and flow rate on the integral behavior of bubble plumes. Journal of Geophysical Research: Oceans, 121(6), 3887–3904. doi:10.1002/2015jc011381

Purpose:

The gas plumes generated in oil blow-outs have an important effect on the oil dispersion and mixing. Due to the difficulties to measure or predict (numerically and experimentally) the near-field dynamics and the further consequences of them, high-resolution modelling is needed.

Data Parameters and Units:

The input parameters for a bubble plume in a tank were included: control.cin, LPT.cin, infodom.cin, mdmap.cin and geom.cin. The model equations work in dimensionless form, but the input parameters are expected in International System units.

Methods:

Hydro3D performs large-eddy simulations on staggered, Cartesian grids. The bubbles are simulated through a home-made Lagrangian Particle Tracking algorithm, using a modified PSI-ball approach for the Eulerian-Lagrangian mapping. Hybrid MPI-OpenMP parallelization is available.

Instruments:

High performance computers and workstations with MPI protocol are advised. Intel FORTRAN compiler was originally used.

Error Analysis:

The model ensures that the mass and momentum conservation laws are fulfilled above a dimensionless error of 1E-11. The model has been extensively validated versus experimental data with remarkable agreement.

Provenance and Historical References:

The model's inputs proceed from experimental datasets. The model was not calibrated for any specific case, using only general formulations from the available literature.

Individual Files: