The densification task in cities makes buildings shoot higher and higher into the sky more and more. To make high-rise buildings as sustainable and self-sufficient as possible, TU Delft PhD student Berk Ekici developed a computer tool that designs the towers at lightning speed using artificial intelligence.
From the risk of anonymity to the higher costs and the uncertain demand from interested buyers: designing the towers in the city creates new dimensions in the design process. But despite these new obstacles, high-rise buildings are still the most used and most attractive way to densify (inner) cities. As spatial advisor for urban development in Rotterdam Municipality, Emiel Arends said earlier this year Area development.nu: “If we want to densify the inner city, we cannot avoid going higher.”
At the same time, self-sufficient buildings play an increasingly prominent role in area development. But how do you make a giant tower not only habitable for residents and local residents, but also self-sufficient? Especially when you are constantly dealing with external (weather) influences. Shielding ensures more use of light and heat on the lower floors, while extreme weather conditions must be taken into account for the upper floors.
TU Delft PhD student Berk Ekici devised a method to find the optimal balance between issues such as energy efficiency, home comfort and optimal use of solar panels even before the construction of these towers. “The buildings of the future must be self-sufficient in electricity, food, heat and water,” says Ekici on TU Delft’s website. “To achieve this, we have to design buildings in such a way that they make optimal use of the available daylight and that they can harvest rainwater and energy themselves. Therefore, we must design better. Artificial intelligence can help us adapt every aspect as efficiently as possible.”
From 17 years to a few weeks
In his dissertation Towards self-sufficient high-rise buildings: Performance optimization using artificial intelligence Ekici explains what the added value can be by using artificial intelligence (AI) in the development of self-sufficient high-rise buildings. The computer links parameters such as the building’s shapes, floor heights and overhangs to the building’s performance. The algorithm can calculate the optimal composition through these combinations.
In addition, each simulation ensures that the algorithm adjusts the search behavior and an even better solution can be selected from among the billions of possible design solutions. “If you had to do it using simulations, it would take about 17 years of work. An AI algorithm can do it in a few weeks.”
Ramp on the roof
Together with his colleague Cemre Cubukcuoglu, Ekici developed the Optimus optimization tool to quickly review billions of design solutions. That in Rotterdam he used as a baptism of fire for Optimus. Based on light and other external (weather) conditions, he divided the three towers into nine separate parts, from ground level to the roof.
The computer eventually calculated the nine optimization problems. For example, if Optimus was in charge, the current uniform surface of the building would disappear. A varied facade design ensures that all nine parts can be utilized optimally and can be protected against external factors. A slope on the roof ensures a higher yield from the solar panels.
Selection of the best
The optimization tool also deals with the interior of the building. Solar panels will of course be placed high on or on the building, but for other functions space needs to be placed lower in the towers because artificial light is used there. “If 1800 people live in the three towers, two courtyards can provide 66 tons of lettuce according to the algorithm and 47.2 percent energy self-sufficiency.”
But, Ekici concludes, despite all these design advantages, the transparency that comes from this way of working is perhaps the most important benefit of his work. “AI allows us to choose the best from hundreds of thousands of design options. It’s a big step forward in performance optimization of self-sufficient high-rise buildings.”