Lausanne, Switzerland — Evaluating the livability of a building after an earthquake is a challenge and a great responsibility for engineers. Beyond the risk of collapse, these experts must predict whether the already damaged building will withstand a similarly violent response in the same place. And in this area, research is still in full swing.
Researchers at the Federal Polytechnic School of Lausanne’s (EPFL’s) Laboratory of Computer Science and Mechanics applied to Construction (IMAC) are proposing a new methodology that can bring more precision to this evaluation. Their approach, based on the vibration analysis of buildings, aims to complete the current tools and to accelerate the sorting between habitable and risky buildings. Yves Reuland (first author), Pierino Lestuzzi and Ian FC Smith published their study in the January issue of Soil Dynamics and Earthquake Engineering.
“In this article, we show that it is possible to bring together existing technologies that have not been gathered so far — the interpretation of the usual metrics used to judge the state of bridges, which we have applied here to the valuation of buildings damaged by an earthquake,” explains Pierino Lestuzzi, master of teaching and research. “The other novelty of our methodology is that we do not need to know the zero state of a building to make an observation.” The researcher indicates that it is still rare for a building to be equipped with sensors that continuously record its behavior. After an earthquake, the engineers therefore make a diagnosis like a doctor who does not have access to a patient’s medical history.
This diagnosis is made by visual observation following a form developed by Italian engineers. This document proved its relevance after the two strong earthquakes that occurred in 2009 and 2016 in central Italy. While a building visit and damage survey are essential after an earthquake, the process is lengthy — it takes 2-3 hours per building — relatively subjective and complex. This finding also entails considerable uncertainty as to the resistance of a building to future replicas. This is where the methodology developed by the IMAC Laboratory comes into play.
Specifically, engineers record ambient vibrations (generated by wind and human activities such as road traffic) from each building using a portable seismograph. They place for this purpose for half an hour three to four sensors in different parts of the building, as a doctor would auscultate a patient using a stethoscope.
Encrypted information and easy sorting
These data are then cleaned to distinguish earthquake-related changes in structure behavior from those caused by weather, ambient noise, or building age. Finally, the researchers enter the parameters obtained in physical models in order to predict the resistance of the building to the replicas. The study estimates between 50 and 100% the degree of reliability of this prediction. These numerical results will complete the estimates made by the visual observation.
“The combination of these processes will reduce the degree of uncertainty about the livability of a building, but developments of our models are of course desirable before they are generalized,” says Yves Reuland, postdoctoral fellow.
Pierino Lestuzzi indicated that the IMAC Laboratory methodology would already be useful for a rapid assessment of buildings located in the outer corona of an earthquake to allow its inhabitants to quickly return to their homes.
For this study, IMAC researchers worked theoretically and experimentally analyzing the destruction of a building and the experimental tests carried out on a test building mounted on a vibrating table by EPFL’s Laboratory of Earthquake Engineering and Dynamics of Structures (EESD), directed by Katrin Beyer.