Prediction of the motion of rocks and steel balls in a section of a semi-autogenous grinding (SAG) mill. Colour indicates particle speed, with red the fastest and blue the slowest.

Prediction of the motion of rocks and steel balls in a mill.

Computational modelling of fluid and particulate dynamics

Our mathematicians are developing equations and software to solve critical fluid flow problems for industry and research.

  • 17 November 2007 | Updated 22 May 2013

Computational Fluid Dynamics (CFD) involves modelling fluid and granular flows to solve high-value industrial problems.

Computational modelling is incredibly versatile. Our scientists build simulations to visualise and analyse complex physical flows for systems as diverse as mill grinding, dam breaks or people swimming.

Building simulations helps us understand system processes as well as predict how changes will affect the system.

Our mathematicians are experts in particle-based modelling methods having developed these techniques over decades of research.

Modelling industrial granular flows

CSIRO models industrial flow problems using unique research, algorithms and software.

Production, processing and transport of granular materials like minerals, powders and cereals are of immense industrial importance.

We have developed Discrete Element Method (DEM) codes for studying various granular flow processes. 

DEM is a modelling technique that maps the behaviour of many individual particles including their interaction with other particles. 

This sort of modelling can improve understanding of flow dynamics and lead to improvements in equipment design and operation.

We use DEM to model industrial processes such as:

  • separation and mixing
  • blasting and fracturing
  • bulk material sampling
  • blending and stockpile construction
  • milling, grinding and crushing
  • mould filling

For example, we are developing virtual grinding mills to improve energy efficiency, improve processing and accelerate machine design.

Smoothed particle hydrodynamics

Smoothed Particle Hydrodynamics (SPH) is a particle-based modelling technique that can track where fluid or will flow. 

SPH has several advantages such as accurately modelling how the fluid surface behaves and taking into account hard to model effects such as wave motion, bubbles in the fluid, splashing and turbulence.

CSIRO scientists use SPH in many fluid flow situations for example:

  • extreme flow events such as dam breaks, landslides, storm surges or volcano eruptions in order to better understand their consequences
  • biomechanical processes such as swimmers moving through water to improve stroke technique
  • digestion processes such as chewing, stomach filling or bowel function
  • casting and moulding applications

SPH’s ability to accurately represent fluid behaviour like bubbles, wetting and foaming means it has even been used to create special effects.

We have created animations of fluids like water, smoke, gases, lava and molten metals for films and computer games.

SPH is traditionally used for fluid mechanics, but is also very useful for solid mechanics where there is fracturing, shattering or possible phase change.

Finite element and finite volume methods

Finite Element Method (FEM) is a technique for solving partial differential equations (PDEs) – equations that involve several interacting variables.

Our software package, Fastflo, uses FEM to solve problems that are best tackled using PDEs.

Fastflo can be applied to one, two and three-dimensional problems encountered by mathematicians, academics and engineers.

The PDEs can be well known, such as fluid flow, heat conduction and electromagnetism problems, or non-standard equations encountered in scientific or industrial applications.

The finite volume method is also used in conjunction with SPH and DEM tehcniques to model complex industrial and geophysical flows.

Read about our work in Process magazine (Oct 07).

Find out more about our work in computational modelling:

Modelling dam breaks, tsunamis and other geophysical events

CSIRO mathematicians are creating computational models of events like floods, dam breaks and tsunamis to aid understanding and planning for these phenomena.

Making movies flow

Take an umbrella next time you go to the movies! CSIRO mathematicians are solving one of the last big challenges in animation: fluids.

Maths for human performance modelling: from speedy swimmers to somersaulting divers

Our expertise in human performance modelling has applications in many sports. Find out how our virtual swimmers may give Australian athletes the edge in the 2016 Rio de Janeiro Olympic Games.

Fastflo: fast, flexible software for mathematicians, scientists and engineers

Our Fastflo software has many applications such as modelling air turbulence, understanding cooling processes or simulating pressure changes.

Modelling industrial granular flows

We’re using computer modelling techniques to model the flow of granular materials such as pharmaceutical or food powders and mineral sands or ores.

Computational modelling of tsunamis and volcanos

How do you know if a natural disaster will destroy your home? In this video podcast, find out how CSIRO technology can map the potential path of a tsunami or volcanic explosion. (1:18)