Since Cali, Colombia is in a seismically active zone, it is essential to look for the best mathematical models to predict how to avoid seismic damage to structures.
It is essential to mitigate seismic risk in new and existing infrastructure in the country, because almost 83% of Colombia's population lives in areas of intermediate and high seismic hazard, according to the Instituto Distrital de Gestión de Riesgos y Cambio Climático.
Oscar David Hurtado, a Master's student in Engineering, Emphasis in Civil Engineering at Univalle, linked to the Research Group in Seismic, Wind, Geotechnical and Structural Engineering (G-7) combined his expertise in programming with his knowledge in structural engineering, making use of data from an international dynamic test to better feed his mathematical models.
In the scientific article entitled Bayesian Model-Updating Implementation in a Five-Story Building published in the international scientific journal Buildings, researchers from Univalle, including engineer Hurtado, adjusted the parameters of a full-scale model of a five-story building and quantified the uncertainty associated with them.
"Mathematical models often use ideal conditions, but in the real world there is a lot of uncertainty; using experimental data, we can incorporate this new information and anticipate the behaviour described by the model by taking this variability into account," Hurtado said.
Photo: Five-storey test building at the University of California, San Diego, with the bare structure without non-structural elements (left) and the complete structure (right).credit: Hurtado et al. (https://doi.org/10.3390/buildings13061568)
The Research
At the University of California, San Diego, in southern California in the US, a full-scale, multi-sensor, five-storey building was built on top of a setup that can simulate an earthquake.
Hurtado and his collaborators had access to the data generated by this test and were able to update their models to be more accurate and robust.
According to Hurtado, every building has a particular frequency or vibration, which is called "natural frequency", and so does every earthquake.
"The frequency coincidence between the earthquake and the building would be dangerous," Hurtado said.
Professor Albert Ortiz, current director of the School of Civil Engineering and Geomatics and a co-author of the scientific publication explained that the main contribution of the project is that from the average response of a building researchers can find properties of the structure without using destructive techniques.
"It is a fundamental input when defining risk issues: if we can measure the response of an occupied building, we can define properties that are very important for predicting its behaviour before future events," said Prof Ortiz.
Daniel Gómez Pizano, senior lecturer in the area of structures at the School of Civil Engineering and Geomatics (EICG) and another co-author of the scientific publication explained that the project stems from a research topic as Hurtado's undergraduate degree project.
"The topic of the thesis was "Structural model fitting using Bayesian inference", a topic of great interest and transversal to the research areas of Professor Ortíz and myself," Professor Gómez said, adding that his main contribution was the contribution of knowledge of dynamic analysis and mathematical modelling of structures.
"Characterising the uncertainty associated with such important parameters in the design of structures as the mechanical performance of their component elements is very important when calculating how confident we are of structural safety in the face of seismic events or other design loads," Professor Gómez said.
Professor Gómez says that in addition to contributing to damage analysis or structural pathology in general, it can be very useful in the stochastic design of structures, which could take into account, for example, variables associated with uncertainty due to human impact on construction.
|
Read more from the School of Civil and Geomatics Engineering: Univalle is Leading Research and Innovation in Reducing Seismic Risk |
The Future
For Hurtado, the biggest goal is to contribute models that can be used to update Colombia's construction standards, which are currently based on US standards.
"The perspective with the incorporation of stochastic modelling in structural analysis is to incorporate the variability inherent in the process of design, experimentation and evaluation of structures into the Colombian seismic-resistant standard, which, being largely based on international construction standards, does not incorporate the uncertainty associated with Colombian materials, construction techniques or design processes," Professor Gómez said.
According to Hurtado, a group of Univalle undergraduate students are already looking for Colombian data on local building materials and conducting more experimental tests in order to add their properties to the model.
If you are interested in contacting the researcher or learning more about the project, please write to the Communications Office Faculty of Engineering: comunicaingenieria(at)correounivalle.edu.co
Cover photo: Oscar David Hurtado. Credit: Alexander Bejarano/NCC-FI/Univalle
Comentarios
Publicar un comentario