Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

Super tall, super smart | Adaptive structures

adapt deflection

Ultra-slim super- tall buildings could be an economic possibility if an alternative “adaptive” design philosophy is adopted.

Any engineer visiting London’s Building Centre recently could hardly fail to have  been impressed by the ultra-slim cantilever on display by the main entrance. With a span/depth ratio of 40:1 and weighing in at just 102kg, the 6m truss is able to sustain a live load of 100kg at its extreme end without visible deflection, thanks to an array of sensors and electric linear actuators built into the structure and controlled by computer.

Given that a tall building is effectively a vertical cantilever carrying wind loads, could the same design principles be successfully applied to ultra-slim buildings, where wind loading is the dominant design case?

Prototype cantelever

University College, London (UCL) research associate Gennaro Senatore certainly believes it can. He points first of all to the prototype cantilever.

“The best span/depth ratio you could achieve with conventional passive alternatives would be 20:1 using steel I-beams, but this would weigh five times as much.

“A conventional truss would be only around 300kg, but its span/depth ratio would be 12:1 at best.”

So the adaptive truss shows massive savings in materials of up to 80%. But it needs electric energy inputs to work, and these have to be assessed in relation to the overall energy inputs to the structure, including the energy embodied in the materials themselves.

Based on the data collected from the prototype, Senatore’s calculation show total structural energy savings of 70% compared to the I beams and 40% compared to the conventional truss.

Key to such dramatic savings is the fact that the actuators only kick in when the loads on the truss produce substantial deflections. Under dead load alone, or with moderate live loads, the structure functions passively.

The Gherkin London

The Gherkin London

The researchers applied their adaptive models to the Gherkin skyscraper, London.

Switching the actuators off with a 100kg load on the extreme end results in a deflection of 170mm. “People often ask me what would happen in a real structure if there was a power cut during an extreme event,” Senatore reports.

“The answer is that in that unlikely event the actuators would lock and the structure would still stand – although there might be substantive deflections, the load carrying capacity would remain.”

And he points out that the prototype truss with actuators on has virtual “infinite stiffness”, that is,  zero deflection under load.

Joint effort

Development of the prototype was a joint effort between UCL and London-based structural consultancy Expedition Engineering. Funding came from the ICE and the Institution of Structural Engineers, along with the Engineering and Physical Sciences Research Council (ESPRC).

Senatore says he first started thinking about the potential of such structures nearly eight years ago.

He adds: “It wasn’t a new idea, the concept has been around since the 1960s, although the focus has been mainly on facades.

“But adaptive structural skeletons have only become possible recently thanks to advances in technology.”

Diagram 2 jpg

Diagram 2 jpg

The adaptive philosophy was shown in case studies to have huge energy savings over traditional materials.

At the heart of the adaptive structures approach is an alternative to the structural compromises needed by the conventional design philosophy.

Any building or structure has to be designed to cope with extreme events, such as high winds, heavy snow loads, unusual crowd movements and the like. The conventional approach, the “passive” approach, is to design structures to be strong enough to resist extreme forces and stiff enough to do so without excessive deformation.

The question is: can we find a better way of controlling movement than simply adding more steel, concrete or timber?

Gennaro Senatore, UCL

But these extreme forces rarely if ever occur. So for the vast majority of its working life a conventional structure is significantly understressed. The “emergencies-only’ structural content is usually an expensive investment, but essential, as so far there has been no practical alternative.

“In practice, more material is usually added to the structure to control movement than to resist loads,” says Senatore.

“The question is: can we find a better way of controlling movement than simply adding more steel, concrete or timber?”

His answer is to separate out the two structural functions. Structural safety is down to the passive component, which is strong enough to resist all predictable loads without collapse. An active combination of sensors and actuators prevents excessive movement under abnormal loads. The key objective is to optimise the balance between active and passive systems to achieve the most effective and economic solution.

Senatore has developed a design method based around the concept of the material utilisation factor, known as the MUT. This is basically the relationship between the maximum likely stresses in the structural elements and the actual capacity of those elements.

Hero image

Hero image

The adaptive structure exhibit ran from 2 August to 1 September at London’s Building Centre, with live demonstrations three times a week.

“A structure with a high MUT would be very light, but would need frequent energy input from the actuators,” Senatore explains. “We have come up with a design algorithm which analyses in detail the effect of varying the MUT.

“An optimised design would achieve minimum whole life energy demand, it would be significantly lighter, quicker and easier to construct and could be a lot slimmer than conventional passive structures.

The algorithm calculates the structural element sizes and the size and location of the actuators for a range of MUTs until it reaches the optimal solution. Stiffness is achieved through the actuators effectively changing the shape of the structure, rather than through structural mass.

Screen shot 2016 09 13 at 14.15.24

Screen shot 2016 09 13 at 14.15.24

“On the prototype truss, the actuators actually create a shallow arch of the bottom chord to cope with the live loads,” Senatore reports. He also points out that even under significant torsion caused by eccentric loading the deflections in the truss were too small to measure, even with a self-levelling laser.

Several case studies have been carried out by the project team. One of the first was of a notional 100m tall building with a planar truss structure subjected to typical tall building wind loading. Where the footprint is very restricted, the adaptive design could achieve a 40:1 slenderness ratio and limit total building drift to under height/500. That is, the highest floor would deviate no more than 200mm from the vertical wind loading.

There was a massive 60% saving on material content, while whole life energy savings on the structural frame were calculated as 40%. “This implies that ultra-tall buildings can be squeezed into restricted city centre plots and still make economic sense,” says Senatore.

“We also found that on such restricted plots, the adaptive high-rise outperforms the passive structure under normal day-to-day loading, not just for extreme cases.”

Adaptivestructures 1 jpg

Adaptivestructures 1 jpg

University College, London (UCL) research associate Dr. Gennaro Senatore programs the truss.

Very slim buildings usually have to incorporate some form of damping system to keep wind-induced sway down to levels that occupants with sensitive stomachs can tolerate. Better aerodynamic detailing of the cladding, optimum orientation with respect to the prevailing winds and a “streamlined” floor plan can help, but extra damping is almost always essential.

But an adaptive structure could need no additional damping. “Basically you select a load activation threshold, below which the actuators are inactive,” explains Senatore.

“Generally you tune it so that they are only needed rarely. It depends on your definition of extreme conditions.”

Gherkin case study

Another case study considered buildings such as London’s iconic Gherkin tower, officially known as 30 St Mary Axe. This has no structural core; all loads are taken by its distinctive exoskeleton. Analysis of a simplified model of this building produced more dramatic results, even though the building itself is far from slim.

An adaptive alternative would require 43% less steel than an optimised conventional design. Whole life energy savings on the structural frame were calculated to be 30%.

Even lighter structures could be possible through the use of higher strength materials. “These are often no stiffer than conventional materials so their full potential can’t be realised,” Senatore says.

“But with the adaptive approach we can overcome any deflection issues and make such materials a practical and effective option.”

Adaptivestructures 23 jpg

Adaptivestructures 23 jpg

University College, London research associate Dr. Gennaro Senatore

Extreme possibilities

In December 1974 a roof collapsed at Tehran’s international airport, killing 17 people. The immediate cause was heavy snowfall on the roof – it was later popularly believed that the roof had never been designed for such loading, the (foreign) designers assuming that Iran was a hot, dry country where it never snowed.

Whatever the truth, heavy snow in Iran is rare but not unknown. Calculating the risks of such extreme events is one of the most challenging responsibilities of structural designers.

Some structures, of course, have very predictable “extreme” events. Sports stadiums stand empty for most of the time but have to support tens of thousands of spectators at intervals – rarely for more than a few hours once a week.

Adaptive design would seem to offer significant benefits to such structures, and to many others. One of the UCL case studies looked at a portal frame, similar to those used for many long span roofs.

This showed material savings of 57% and energy savings of 53%. A similar study of a large vaulted roof produced very similar results. Structures where stiffness is the governing factor seem to benefit most from the adaptive approach.

Fewer benefits were apparent when a naturally stiff catenary arch bridge was analysed. Here strength was the governing factor, actuators could contribute less, although a mass saving of 35% was still considered possible.

Slender, long span bridges, be they cable stayed or suspension, would also seem a natural application for adaptive technology, to control movement during extreme wind events. Slashing the materials needed by double figure percentages would have a massive effect on the economics of the crossing – as well as producing some really dramatic bridges.

Readers' comments (1)

  • This seems to be an innovative and rightly praised design theory. My main concern would be the application to combat wind loading on tall buildings. Wind loading is generally a very instantaneous load - especially in the extreme events for which it would be used. How quickly can these structures change shape to resist the extreme force? Would it need localised weather forecasting to determine when it should change shape and for how long, or could it respond as instantaneously as the load is applied to prevent a 'lag time'?

    Unsuitable or offensive? Report this comment

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Please note comments made online may also be published in the print edition of New Civil Engineer. Links may be included in your comments but HTML is not permitted.