Skip to content
Civil Engineer blog

How artificial intelligence can help improve wind resilience of buildings and structures

15 May 2023

RWDI’s Daniel Hackett discusses new approaches to engineering modelling and analysis to mitigate wind impact in a changing climate.

How artificial intelligence can help improve wind resilience of buildings and structures
How winds affect the pedestrian environment should be fully analysed at the design stage to avoid costly amendments. Image credit: Unsplash

Mitigating the impact of wind is an important and well-established part of good building design.

We must create structures that can withstand the highest wind loads a building might be subjected to. And, they must move in a way that doesn’t cause discomfort or distress to the people that use them.

We must ensure that cladding elements are strong and secured enough to remain intact and in place despite the pressure exerted on them by the wind.

We also need to create a pedestrian environment that’s safe and suitably comfortable, and that isn’t affected by wind being deflected or accelerated by the nearby buildings.

All these things require the input of an experienced wind consultant applying sophisticated methods and tools, such as large boundary-layer wind tunnels and advanced computational simulations.

The importance of assessment and analysis

These types of assessment are complex.

The results can be sensitive to relatively small architectural features of the building as well as to the nature of the area surrounding the project site.

As such, wind assessments tend to happen late in the design process, when many of the parameters of the design have been fixed.

This reduces the uncertainty in the output and limits the need to conduct multiple rounds of assessment.

Early input can provide crucial benefits

But input from wind consultants at an earlier stage can provide crucial benefits in terms of identifying and avoiding significant wind issues before the design becomes too far advanced to make necessary interventions.

No design team wants to be told, one month before a planning application deadline that their new building will cause wind safety issues for pedestrians and cyclists.

The efforts required to overcome such a problem aren’t insignificant:

  • there might be a need for additional wind tunnel testing to develop a wind mitigation strategy;
  • the expensive implementation of large and undesirable wind mitigation measures;
  • the re-design of other parts of the building to accommodate them; and
  • costly delays to the entire process associated with doing all of the above.

These problems can be limited or avoided altogether if such wind-related issues are identified early and wind mitigation strategies fully integrated as part of the design progress.

Opportunities to reduce carbon

There are other benefits too.

When it comes to the structural and cladding design, there are potentially large savings to be found in material cost and in the building’s carbon footprint.

For example, by reducing the amount of concrete needed for the structure by making an early and accurate prediction of the likely wind loads.

Intelligence, data, and assessment

This is where data-driven predictive tools come in.

By drawing on the results of thousands of previous wind tunnel tests and simulations, a practitioner can arrive at a good understanding of how a given design is likely to perform.

These are made even more powerful by recent advances in artificial intelligence and machine learning.

These can digest enormous amounts of past data and systematically make use of them to provide new and valuable predictions in a time-efficient manner.

When these tools are placed in the hands of experienced wind consultants, it becomes possible to provide that vital early-stage design input quickly and reliably.

They don’t replace the full, quantifiable assessment methods.

Indeed, the importance of having experienced engineers in the process is partly to understand the limitations of the tools and interpret their output in a useful and robust way.

Instead, they go 'hand-in-hand'.

By using the right tools at the right time, a design team can benefit from cost savings and the early identification of critical wind issues - hopefully avoiding any costly surprises.

Learn more

To find out more on this topic, attend the AI and data-driven approaches to early-stage wind design lecture, presented by Joseph Symes and Andrew Proud from RWDI.

This event is organised in partnership with the Wind Engineering Society and will be held on 17 May 2023 from 18:00 – 19:30 BST at the ICE.

Unable to attend in person? You can watch online, or view the recording at a later date.

Register now
  • Daniel Hackett, area practice leader at RWDI