Seismic stress test of built environment

1) Project title: Seismic stress test of built environment 2) Content of research project The objective of this project is to develop a methodology for seismic risk assessment of the built environment based on data derived for individual buildings. The outcomes of this project will serve as a basis for establishing contemporary regulations to strengthen community resilience in response to earthquakes and other natural hazards and to inform stakeholders about the seismic risk through a novel Buildings Seismic Performance Certificate. 3) Hypotheses Only a few strong earthquakes, which hit Italy in recent years (L`Aquila 2009, Emilia-Romagna 2012, Norcia 2016), have caused hundreds of casualties, huge economic loss, social distress and enormous difficulties of community to recover from disasters. Similar earthquakes may occur in other parts of Europe, where renovation costs could quickly exceed community capacity to recover. Current legislation in Europe is not designed in a way that would enhance the community seismic resilience, since the problem of seismic safety of a building and the built environment is not addressed as a coupled problem. It can be, thus, concluded that the current regulation for establishing the community seismic resilience is in conflict with the EU`s Digital Agenda and the guidelines of the Sendai Framework for Disaster Risk Reduction 2015-2030. This issue can be solved with assuring the flow of information from building level to built environment level. 4) Methods Systematic coupling of the two levels of analysis, that is the level of the individual building and the system scale of the build environment, will be achieved through launching a novel building information model (BIM) that will contain key data about seismic analysis results of individual buildings. This model can be used at the level of the built environment to perform physically-based simulations of its seismic response at the system scale. Existing seismotectonic models will be used to simulate ground motions at the considered site. The seismic response analysis at the building level will be based on simplified nonlinear analysis. As a novelty, the results will, however, be presented in a form of a Buildings Seismic Performance Certificate. The proposed probabilistic framework will couple seismic hazard, structural response, structural damage and losses in terms of costs/repair time. 5) Explanation indicating what is new and/or special about the project With the developed methodology it will be the first time possible to simulate seismic response of built environment for a given earthquake, which will be used for realistic seismic risk assessment of built environment. This problem is computationally extremely demanding, but it will be solved by a time efficient computational procedure through systematic coupling of the analyses at the level of a building and the built environment. The outcome of the project has the potential to change the current European seismic regulation in order to establish community resilience against natural disasters.

The project's overall focus was to strengthen the community seismic resilience via a framework for seismic risk assessment which should be practical for individual-building risk estimation as well as for the built environment. Novel procedures were to be investigated to derive estimations of the capacity of such buildings utilizing limited information (i.e. geometry and approximate data on material properties and structural strength). Special focus was given in pushover-based schemes and their ability to account for the effects of higher modes of vibration and stability problems that may encounter during earthquake excitation. In pursuing a robust and efficient methodology for seismic risk assessment of a built environment, it was necessary to investigate the most contributing factor of uncertainty which is the ground motion excitation. The developed ground motion record selection procedure satisfies the requirements of spectral compatibility, hazard and intensity measure consistency, seismological and site-specific criteria through multi-objective optimization and is based on genetic algorithms. It is particularly useful for regional risk assessments of building portfolios since it does not depend on specific building properties, common in other selection strategies. Through this ground motion selection methodology, multi-component accelerograms can be selected to match predefined spectral targets in terms of first and second order statistics in a wide period range. A unique element is the ability to include probability distribution targets in specific ordinates of the spectrum, on top of the mean and standard deviation, allowing for more refined ground motion sets that can be used to reduce the number of records required in response history analyses, thus minimizing the computational cost. For regional risk assessment of building portfolios, alternative solutions that provide reliable collapse risk estimates with reasonable computational time were intended. To this end, the efficiency and accuracy of simple machine learning tools, such as linear regression and logistic regression, for predicting collapse intensity was investigated, aiming to serve as the solution when there is deficient available documentation of the existing buildings. For the development and testing of machine learning algorithms, a structural response database was developed for a steel frame structure by conducting extensive incremental dynamic analyses, employing a total of 17,141 earthquake records. The analyses revealed distinctive patterns of collapse capacities. Regularized regression was employed for feature selection in which the contribution of a variety of intensity measures to the prediction of collapse was examined. The project results have the potential to be incorporated into future codes to improve the resilience of the community to the effects of earthquakes or other natural hazards, as a clear assessment of the seismic risk of the building stock and its communication are overdue. The project partners will continue their research in this area to quickly approach this goal.