- Project plans
- Project activities
- Legislation and standards
- Industry context
Last edited 23 Apr 2019
CFD - Bridging the gap between architecture and engineering
Computational fluid dynamics (CFD) can play an integral role in all areas of building design, providing accurate and time-efficient simulations of building performance relating to air flow, temperature, pressure and other similar parameters.
In this article, we explore the benefits of computational fluid dynamics software as a design assistance tool and identify where it is actively bridging the gap between architecture and engineering, particularly for architects, heating ventilating and air conditioning (HVAC) engineers - and those in the construction sector who wish to better optimise building designs.
Understanding how natural phenomena affect buildings, particularly internal and external airflows, is an increasingly important aspect of architectural design. This is largely due to the increasing complexity of contemporary buildings and a growing interest in improving building performance in terms of the environmental impact (Kaijima et al., 2013).
CFD has proven to be a key factor for performance enhancement in a number of areas from Formula 1 to the development of swimwear, and the benefits experienced by many industries are now being felt by those in the construction sector.
It is ideal for modelling:
- The thermal comfort of occupants;
- Distribution of environmental conditions within a space;
- Effectiveness of natural ventilation (including the stack effect);
- Heat losses through exterior walls or glass;
- Effectiveness of air inlets, extractors, radiators:
- Build-up of heat in key spaces;
- Positioning of sensors to detect heat or cooling;
- Positioning of major HVAC equipment;
- Wind loading and forces imposed on a building, and
- The impact of a new building on air movement around a site.
Traditionally, CFD had many barriers including hefty licenses, large hardware and power requirements, and an intrinsic understanding of the equations involved. Today, vendors have worked to create high performance solutions that can be accessed on-demand, with a pay-as-you-use model, intuitive code and simple graphical user interfaces (GUI).
 Validating the building design process
As the range of computational fluid dynamics applications continues to increase, new techniques have been introduced that facilitate its use in both architectural engineering and HVAC design (Zhang et al., 2009).
The American Institute of Architects (AIA) says design and construction projects typically involve several phases, or the 'five phases of design'. During the first stage, a variety of vital decisions are made and discussions held regarding the requirements of the project. This is a key part of the design process and often involves stakeholders laying out their clear expectations. Once the object has been defined, the next step is to create schematic designs. It is during this phase that study drawings, documents, and other media are created to illustrate the concepts of the design.
It is also during this phase that CFD is often typically employed, calculating airflows in and around the buildings. Based on calculated results, decisions can be made as to whether the design needs to be modified. These steps are often repeated to ensure indoor and outdoor environmental conditions are satisfactory (Glicksman & Link, 2006), and to inform decisions made by architects and designers.
Affordable and accessible CFD tools are providing greater flexibility to architects during the design process, enabling them to simulate designs for validation purposes throughout the whole process, rather than simply the schematic design stage. This can only be a positive development for all involved, ensuring the end design is the very best it can be.
 Preparing for natural weather events
In addition to being responsible for a large number of deaths each year, natural disasters and weather events such as cyclones, earthquakes and landslides have a considerable impact on a country's economy, mostly due to the need for disaster aid and building recovery.
The effects of these weather events on buildings cannot be understated with serious consequences – such as roof failures – not uncommon (Morrison & Kopp, 2011). Aerodynamic loads on the roof and walls of a low building are characterised by the interaction of wind flow with the surface of the building and this interaction depends primarily on the building geometry and flow characteristics (Stathopoulos, 1984). Traditionally, engineers and architects have rarely considered such factors, instead focusing purely on the design or structural stages of a building design such as walls, overhangs, foundation and roofing.
Managing the huge risk to buildings from wind requires a high level of information on the type and magnitude of wind loads likely to be faced. This information has traditionally been gathered through a combination of full-scale measurements and wind tunnel tests that can prove both costly and time-consuming. For this reason, Computational Fluid Dynamics (CFD) methods provide a useful tool for predicting turbulent flow over buildings, informing decisions and influencing design.
As CFD becomes more accessible, architects and engineers have the ability to simulate the flow field around a building and successfully predict parameters of interest including velocity, pressure, and temperature fields. This could contribute hugely to preparing buildings for natural disasters and reducing repair costs in the wake of such occurrences.
 HVAC system optimisation
CFD simulation is important in the design and optimisation of sensitive HVAC environments. It is particularly useful for performance prediction, providing key HVAC design parameter information, validating design parameters, and modifying malfunctioning HVAC systems. A number of industries have specific HVAC requirements with two examples being that of the clean-room and a medicinal cannabis growing facility.
More and more, HVAC engineers are moving to CFD to compute airflow patterns and space temperatures based on complete 3D geometries, resulting in a greater level of accuracy. Examples of HVAC CFD analysis in practice include:
- Industrial ventilation design;
- Swimming pool ventilation;
- General office/room simulations
- Fume hood design;
- Contamination in a sensitive zone;
- Room pressurisation;
- Effective smoke evacuation in smoking lounges;
- Fire and exhaust simulations in tunnels or parking garages;
- Thermal assessment of data centres and server rooms, and
- Smoke and fire propagation simulations and implementation of fire safety in occupant structures.
Fig. 1 - A HVAC Simulation of a Concert Hall]
Wind tunnel testing has been employed extensively for industry and research applications over the past five decades. The debate as to whether CFD or wind-tunnel testing is the best course of action for architects lives on, with both methods providing a certain degree of knowledge and understanding of the environment in which the design exists.
Wind-tunnel testing requires a costly setup and sophisticated instruments to measure a range of field variables (wind velocity, pressure loads, turbulence intensity etc). Its main limitation is that such measurements are obtained at only a few select points within the test section, severely restricting overall understanding of the evolutionary or transient processes of unsteady complex phenomena (such as vortex shredding, turbulence wakes, and thermal stratification). In addition, wind-tunnel testing can often present a dangerous or impractical challenge to those concerned.
According to Wainwright and Mulligan (2004), CFD offers a wealth of advantages compared to wind-tunnel testing. In addition to generating full-scale simulations (rather than scale models of many physical simulations), it also provides more extensive data that can be measured in the lab. Its results can also be visualised clearly. With CFD, you can run large-parameter sets of 50 simulations at a time in a single working day to evaluate a set of different designs or wind angles on the building.
In addition, physical safety concerns and practical limitations are clearly eradicated.
On-demand, pay-as-you-go tools such as EXN/Aero have enabled architects and engineers to access CFD only when they need it, rather than paying for costly licenses traditionally associated with this area of computing. This means consultants can include simulation software in their tender pitches, only paying for it should they win the pitch. Affordable simulation software also enables architects to provide impressive visualisation for potential clients, reduce overall project costs and limit timescales.
 Conclusion: Overcoming limitations
Architects, consultants and engineers in the building performance field have turned to CFD design and engineering techniques due to the detailed information provided through the design process. The flexible and interactive design environment lends itself to the design decision-making process, just as it has in the aerospace and automotive sectors.
Affordable, accessible CFD applications have encouraged more architects and non-CFD experts to embrace simulation. While this has many positives, a number of which are addressed in this article, some concerns remain about user knowledge and experience, particularly where simulation results inform vital decisions.
Vendors are addressing these concerns by creating intuitive platforms that are generally easy to use, and providing training and support to educate and inform those wishing to benefit. It is not practical for architects and those in the building profession to become extensively trained in the intrinsic particulars of CFD engineering, and vendors must develop code and solutions that combine high performance credentials with an efficient learning protocol.
 Related articles on Designing Buildings Wiki
- Computational fluid dynamics for buildings.
- Computational fluid dynamics in building design: An introduction FB 69.
- Indoor air velocity.
- The design of temporary structures and wind adjacent to tall buildings.
- Thermal behaviour of architectural fabric structures.
- Wind load.
 External references
- Adamu, Zulfikar A., Malcolm J. Cook, and Andrew DF Price. “Natural Personalised Ventilation-A Novel Approach.” International Journal of Ventilation.
- Glicksman, L., & Lin, J. (2006). "Sustainable Urban Housing: Alliance for Global Sustainability".
- Kaijima, S. (2013). "Computational Fluid Dynamics for Architectural Design".
- Kim., D. (2013. "The Application of CFD to Building Analysis and Design".
- Morrison, M., & Kopp, G. (2011). "Performance of connections under realistic wind loading".
- Stathopoulos, T. (1997). "Computational Wind Engineering: Past Achievements & Future Challenges".
- Wainwright, J. & Mulligan, M. (2004). "Environmental Modelling".
- Zhang, L.P., & Wang, Z.J. (2004). "A block LU-SGS implicit dual time-stepping algorithm for hybrid dynamic meshes".
Featured articles and news
Five things to consider before installing solar panels.
New conservation building for the Louvre completed.
A balance between character and climate.
Bamboo pavilion built at London South Bank Uni.
Bringing in an expert.
Why the lowest price isn't sustainable.
The Most Economically Advantageous Tender.
Pipe dream or possibility?
The New Rules of Measurement.
Prioritising Sustainable Development Goals on projects.