Computational Fluid Dynamics (CFD) is a stream of fluid mechanics that utilises numerical methods to analyse and solve problems involving fluid flows, write Jennifer Keenahan and Réamonn Mac Réamoinn.

Navier-Stokes equations form the fundamental basis of these problems. While the use of CFD is commonplace in the automotive and aerospace industries, its application for civil engineering applications is cutting edge. CFD is a powerful tool that adds valuable information to fluid flow problems.

Computation Fluid Dynamics may be used to:

  • HVAC performance analysis: CFD simulations can predict 3D flows in and around buildings. As a design optimisation tool, it allows detailed prediction of local air velocities, air temperatures, surrounding surface temperatures as well as occupant comfort for critical spaces.
  • Data centre air management: CFD analysis provides a detailed view of the actual efficiency, possible optimisation and management of cooling operations.
  • Solve fire and smoke problems: where simulations can predict how the fire spreads and how the temperature rises in enclosed spaces.
  • Jet impulse fan system design: CFD assists the understanding of air movement and smoke behaviour to select the number, size and position of jet fans to achieve the most cost-effective solution in accordance with fire safety codes.
  • Wind engineering: CFD can address design issues related to the environmental distribution of pollutants, ensuring that the dispersal of noxious gases is within the environmental limit.
  • Facade design: fluid dynamic simulations can be used to analyse the physical phenomena occurring in ventilated facades, it can optimise design of ventilation air gaps and openings.
  • Aero-acoustic: flow noise can be simulated coupling aero-acoustic analogies and method with CFD. Tonal noise from building facades is the indicator of a physical phenomenon known as ‘Aero-acoustic resonance’ – using computational air flow assessments it is possible to quantify the potential extent of wind noise generation.
  • Micro-climate and environmental analysis: simulations allow for assessment of outdoor comfort levels in urban areas, including shading canopies, wind barriers and vegetation areas.

CFD is the use of computers and numerical methods to solve problems in fluid flow. It is a method for solving partial differential equations in continuum mechanics using numerical techniques.

It involves breaking problems down into discrete numbers of volumes, thus making them easier to solve. Combining the results of these small volumes permits the generation of the overall solution.

The equations governing fluid motion are based on the fundamental physical principles of the conservation of mass, momentum and energy.

Modelling in CFD comprises three main stages: pre-processing, simulation and post-processing. Pre-processing firstly involves the construction of the geometric model for the flow domain of interest, and the subsequent division of this domain into small control volumes (cells).

‘Meshing’

This process is often called ‘meshing’. The flow field and the equations of motion are then discretised, and the resulting system of algebraic equations is solved to produce values at each node.

Once the model and the mesh have been created, appropriate initial conditions and boundary conditions are then applied.

The Navier-Stokes equations, the governing equations for the behaviour of fluid particles, are solved iteratively in each control volume within the computational domain until the solution converges.

The field solutions of pressure, velocity, air temperature and other properties can be calculated for each control volume at cell centres and interpolated to outer points in order to render the flow field.

Predicted flow field

Post-processing involves plotting the results and visualising the predicted flow field and other parameters in the CFD model. The Navier-Stokes equations, used within CFD analysis, apply a numerical representation to approximate the laws of physics to produce extremely accurate results, providing that the scenario modelled is representative of reality.

In the advent of improving computational power, and the development of numerical techniques such as Finite Element Analysis, CFD offers an opportunity to model many variations of the same problem at full scale, with increased efficiency, in a virtual environment.

In order to capture the phenomena in sufficient detail, large finite volume models are necessary, requiring significant processing power. These large models generate a substantial amount of data, which, in turn, needs to be stored and analysed.

The advantages of CFD, however, far outweigh the disadvantages. It is a non-intrusive, virtual modelling technique with powerful visualisation capabilities.

Results can be captured across the entire domain. There are also significant cost and time savings with CFD, as it makes it possible to assess comparisons between alternative systems quickly and efficiently, without the disruption of making physical changes on site.

As with any computer simulations, the quality of the results is dependent on the quality of the inputs, assumptions, modelling characteristics employed and equations used to represent the phenomena. There will inevitably be approximations; hence, a robust model validation process is essential.

A high level understanding of the modelling process and of the phenomena being modelled is necessary to ensure that the output will be of practical use.

Authors: Jennifer Keenahan, assistant professor and Chartered Engineer in civil engineering at UCD. She graduated with a BE in Civil Engineering from UCD in 2011 and a PhD in 2014. She was a member of the bridge team and the CFD team at Arup from 2014 to 2017 before taking up her current role in UCD as assistant professor in engineering. She is now leading a research team in the area of CFD for the built environment.

Réamonn Mac Réamoinn, wind specialist and Chartered Engineer in Arup. He graduated with a BA BAI in Civil Engineering from TCD in 2005 and later received the CRH-Fulbright award in 2013 to complete a master's in wind engineering (MSE) at Johns Hopkins University in the US. He leads the CFD team in Dublin. His interests include wind microclimate studies, bluff body aerodynamics, fluid-structure interaction, thermal plumes, pollutant dispersal, statistical and probabilistic analysis. In particular, he is seeking to better understand these phenomena and their effects through CFD.