In Scotland alone one historical building is destroyed each month by fire, or at least very badly damaged. This means a huge cultural loss as such buildings can rarely be restored. Within a European research project (1) institutions from twelve different countries are now investigating how fires in important buildings can be caught and combated in time using minimally invasive techniques. "The main goal of the project is to check the incessant destruction of valuable cultural assets", says Alfred Moser, head of the group Air & Climate (2)at the Institute of Building Technology at ETH Zurich, who is also co-ordinating Swiss participation in the project. "In buildings that are open to the public it is also a question of ensuring the safety of the visitors."

Conference venue as case study

How this can be achieved in practice was the subject of a discussion among participants at the beginning of December in a seminar at the palace of Schönbrunn in Vienna. Participants made the venue of the seminar itself an object for a case study. In the case of a fire in this vast palace, not only would valuable substance have to be preserved; visitors would also have to be rapidly evacuated. In the seminar investigations were undertaken to determine in how far numerical simulations can be used to evaluate various fires scenarios and preventative measures. "With our aero dynamical analyses we can show how smoke spreads inside a building and how long emergency exits would remain passable," explains Moser.

Calculations show time and again that even relatively simple measures can play an important role in influencing the way a fire develops. A simulation of a fire in the former Schiffbauhalle in Zurich – which has been turned into a theatre – shows, for example, that the risk to visitorsin the foyer is a lot higher when the connecting door giving access to the big hall is closed. In this particular case the foyer is filled with smoke within minutes. Visitors would hardly have time to escape. If the door is open, however, even after ten minutes the air in the lower part of the foyer is not filled with smoke and there is therefore more time to evacuate visitors.

A draught in the sports stadium

"Great progress has been made in recent years with flow simulation," explains Daniel Gubler from the company Air Flow Consulting (AFC) in Zurich (3). "Notably complex geometries and forms can be technically simulated in models with a reasonable level of resources nowadays." AFC, a spin-off company, arising from ETH’s Air & Climate Group, offers its clients flow analyses, as developed at ETH, as a commercial service.

Possible applications are by no means limited to fires. For instance, AFC also examined how the aeration in the chamber used by Chur’s parliament could be improved. A completely different direction is followed with the simulation of the wind conditions in the new sports stadium planned for Zurich. In this spectacular project a horizontal interstice is planned between the pentagonal shaped tribune and the foundation of a shopping centre. Because of this, at high winds, the stadium could be filled with the troublesome manifestations of draughts. "With our simulations we can devise the best ways of preventing this," says Gubler.

Greater expectations

"The progress referred to could not have been achieved without the huge increases of hardware perfomance that have taken place," says Alfred Moser. "Although that doesn’t mean that computations take less time today. This is because expectations have grown in line with technical developments. We are constantly trying to integrate new aspects into the models so that computations are as realistic as possible." Asked where further research is needed Moser says that above all, in the modelling of turbulences. "Unstable layers and the accompanying turbulent currents also occur, for instance, in rooms with floor heating," and he goes on to explain: "Warm air from the floor does not rise evenly over the entire room. Better models are needed in order to replicate such processes more accurately."

Computer simulation models can help to show from where the wind will blow in Zurich’s new stadium. (Picture: D. Gubler, AFC) large

A relatively new area of application for the team is that of cleanrooms in hospitals. "A current application arose in connection with the SARS crisis," says Moser. The University Hospital had to furnish isolation rooms, where suspected cases could be investigated until more was known, with mobile filter equipment, so that the virus could not infect nursing staff or other patients or be carried to other rooms. "We measured how quickly the equipment filtered the air in the isolation rooms and whether air pockets occurred that could harbour viruses," he explains.

How precise does the model need to be?

Moser also sees possibilities for the designing of operating theatres. "We calculate whether only clean air reaches the operating table with a certain type of aeration system. "One major factor in modelling such a situation is the problem of how to simulate the movements of the operating surgeon. Is it sufficient to represent the operating team in a simple cylindrical formation around the operating table or do arms working above the patient also have to be taken into account?

This example shows that carrying out calculations is only one aspect of the problem. "The first challenge for the engineer is to decide which method to employ," explains Daniel Gubler. How does one build a model, which scenarios are really relevant for the evaluation, and on the basis of what assumptions must one proceed? All these questions – and others – must be addressed and answered before the computer can start its work. "A lot depends on the assumptions one includes, especially with the simulation of fires," says Gubler. "And these uncertainties absolutely have to be taken into account when it comes to interpreting the results."